<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Orphéric ALLAGBE on Medium]]></title>
        <description><![CDATA[Stories by Orphéric ALLAGBE on Medium]]></description>
        <link>https://medium.com/@Orpheric_Codeur?source=rss-ba8df5dc9a11------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*blKqzzmhtygbjILajEJT7w.jpeg</url>
            <title>Stories by Orphéric ALLAGBE on Medium</title>
            <link>https://medium.com/@Orpheric_Codeur?source=rss-ba8df5dc9a11------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 02:46:20 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@Orpheric_Codeur/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Modernisation des infrastructures de santé : migration d’un système médical sur site vers le Cloud…]]></title>
            <link>https://medium.com/@Orpheric_Codeur/migration-vers-le-cloud-du-syst%C3%A8me-dinformation-du-centre-hospitalier-international-de-z%C3%A9m%C3%A9-cas-de04a3a8f0ae?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/de04a3a8f0ae</guid>
            <category><![CDATA[database-migration]]></category>
            <category><![CDATA[microsoft-sql-server]]></category>
            <category><![CDATA[aws-database]]></category>
            <category><![CDATA[healthcare-technology]]></category>
            <category><![CDATA[cloud-migration-strategy]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Tue, 10 Nov 2026 14:21:03 GMT</pubDate>
            <atom:updated>2026-11-10T14:56:41.174Z</atom:updated>
            <content:encoded><![CDATA[<h3>Modernisation des infrastructures de santé : migration d’un système médical sur site vers le Cloud d’AWS; Cas du Centre Hospitalier International de Zémé<strong>*</strong></h3><p><strong>Contexte Technique :</strong> Le Centre Hospitalier International de Zémé souhaite moderniser son infrastructure sanitaire à travers sa solution de planification des rendez-vous médicaux, en migrant l’ensemble de son système de base de données <strong>Microsoft SQL Server</strong> vers <strong>Amazon Aurora</strong> pour réduire les coûts de base de données commerciales, tout en améliorant la fiabilité et les performances du SI.</p><p>La transition d’une base de données <strong>Microsoft SQL Server</strong> vers <strong>Amazon Aurora </strong>est une excellente décision, et <strong>AWS </strong>dispose de plusieurs outils qui peuvent aider à rendre ce processus plus fluide et plus efficace.</p><p>A cet effet, les services <strong>AWS SCT </strong>(<strong>Schema Conversion Tool</strong>) et <strong>AWS Database Migration</strong> (<strong>DMS</strong>) seront respectivement utilisée pour convertir le schéma et le code de la base de données source en un format compatible et migrer les bdd relationnelles concernées.</p><p>Concrètement, <strong>AWS SCT</strong> peut nous aider à analyser notre système de base de données <strong>Microsoft SQL Server</strong> existante, identifier les problèmes de compatibilité, puis générer les scripts nécessaires pour convertir le schéma dans un format compatible avec <strong>Amazon Aurora</strong>. Cela garantira une migration fluide de la structure de notre base de données.</p><p>D’autre part, <strong>AWS DMS</strong> sera être utilisé pour effectuer la migration réelle des données de notre base de données Microsoft SQL Server vers Amazon Aurora. C’est un service <strong>entièrement géré</strong> qui peut gérer le transfert de données avec un temps d’arrêt minimal, garantissant que notre système de planification de rendez-vous médicaux reste opérationnel tout au long du processus de migration.</p><p>En matière de <strong>coût</strong>, de <strong>fiabilité </strong>et de <strong>performances</strong>. <strong>Aurora </strong>est conçu pour être plus rentable que les solutions de base de données traditionnelles, avec jusqu’à <strong>90% d’économies par rapport aux bases de données commerciales</strong>. Côté fiabilité, à titre d’exemple, il fournit <strong>six copies de nos données sur trois zones de disponibilité (AZs)</strong>, garantissant une haute durabilité et disponibilité.</p><p>Particulièrement pour les <strong>performances</strong>, Aurora utilise les dernières technologies matérielles et de stockage pour offrir des opérations de base de données à faible latence et à haut débit. Il fournit jusqu’à <strong>5 fois le débit</strong> des bases de données MySQL et PostgreSQL standard.</p><p>Pour mieux expliquer le processus suivi, voici les détails sur les étapes</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/976/1*OInvp9S52NhkKxC9zdj0xA.png" /><figcaption>Architecture cloud du système conçu via AWS SimuLearn</figcaption></figure><p><strong>Etapes 1 &amp; 2 :</strong> Une application pour planifier des rendez-vous médicaux est hébergée sur un serveur web situé sur les infrastructures sur site (datacenters), avec une base de données Microsoft SQL servant de backend</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZAIdu0edKIuGzHK9RNS9uw.png" /></figure><p><strong>Etape 3:</strong> L’outil de conversion de schéma AWS (AWS SCT) est utilisé pour transformer le schéma de base de données actuel du moteur Microsoft SQL vers le moteur PostgreSQL.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iEeE1NU-4PsQjwmtIYPabA.png" /></figure><p><strong>Etape 4:</strong> Ici, un nouveau projet est créé, et une connexion est établie, entre <strong>AWS SCT</strong> et à la fois la base de données source, <strong>Microsoft SQL</strong>, et la base de données cible, <strong>Amazon Aurora</strong> <strong>PostgreSQL</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yX5-89Z3Hr-oJo2Ez5nbMg.png" /></figure><p><strong>Etape 5 &amp; 6:</strong> <strong>AWS SCT</strong> convertit le schéma au format requis pour la base de données cible, et il génère des rapports d’évaluation de migration complets.</p><p><strong>NOTE IMPORTANTE:</strong> Avant de convertir le schéma avec AWS SCT, des règles de migration peuvent être établies. Ces règles peuvent modifier le type de données des colonnes, relocaliser des objets entre les schémas et modifier les noms des objets.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*j7Bjb8TqyUfdjzGGEJPzMw.png" /></figure><p><strong>Etape 7:</strong> Les éléments du schéma <strong>scheduling_db </strong>de la base de données source sont convertis en schéma équivalent pour le moteur de base de données de l’instance de base de données cible.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*O7904MA-hYxjr0RYL0Dufg.png" /></figure><p><strong>Etape 8:</strong> Une tâche <strong>AWS Database Migration Service</strong> (AWS DMS) est créée pour migrer la base de données en utilisant une instance de réplication AWS DMS et des points de terminaison.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pFmnc4UYsPW9ouA6v9QCEA.png" /></figure><p><strong>Etape 9: </strong>Une fois la connexion avec les points de terminaison établie, la <strong>tâche de migration</strong> est lancée et la progression est suivie à l’aide d’<strong>AWS SCT</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZsFSbmtIIyA2DSkEWAnjMA.png" /></figure><p><strong>Etape 10: </strong>Après la migration des données, la base de données cible est soigneusement testée pour s’assurer que les données ont été correctement migrées et que la base de données cible fonctionne comme prévu.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3WgBfGAg6V7GXf4QfUYK5g.png" /></figure><p>Ici, au 11, l’application est modifiée pour utiliser la base de données cible, et des tests sont effectués pour s’assurer que l’application fonctionne correctement.</p><p>Pour finir, La même méthodologie peut être appliquée pour <strong>migrer la base de données Microsoft SQL vers la base de données Amazon Aurora MySQL.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BS57zoj3Ltz3PeWYN_FZqw.png" /></figure><p>Maintenant, essayons d’implémenter tout ceci:</p><p>Dans les étapes suivantes de notre implémentation, nous allons dans un premier temps démontrer comment utiliser <strong>AWS SCT</strong> pour convertir les schémas de bases de données afin de faciliter la migration de base de données et ensuite démontrer comment utiliser <strong>AWS DMS</strong> pour effectuer des migrations de bases de données hétérogènes.</p><p>Pour cela, voici les taches à accomplir :</p><ul><li>Examiner la base de données et l’application qui ont été déployées sur site.</li><li>Générer une clé d’accès et une clé d’accès secrète<strong>.</strong></li><li>Transformer le schéma “<strong>scheduling_db</strong>” de MS SQL Server en format compatible PostgreSQL en utilisant AWS SCT.</li><li>Migrer la base de données et les tables de <strong>MS SQL</strong> <strong>Server </strong>vers un <strong>cluster Aurora Serverless</strong> en utilisant AWS SCT pour la compatibilité PostgreSQL.</li><li>Mettre à jour les données utilisateur de l’application avec les nouveaux paramètres PostgreSQL et vérifier la migration de la base de données dans l’interface de l’application.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ChanMdjQ_RvbA2CCk6918A.png" /><figcaption>Architecture récapitulatif : Migration de base de données d’une app de rendez-vous médicaux vers le cloud d’AWS</figcaption></figure><p>*<strong>Note:</strong> Afin de respecter la confidentialité, le nom <strong>Zémé</strong>* du “Centre Hospitalier International” en question, est purement fictif.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=de04a3a8f0ae" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Une année de plus en career cloud : The Recap’]]></title>
            <link>https://medium.com/@Orpheric_Codeur/une-ann%C3%A9e-de-plus-en-career-cloud-the-recap-d9e7d8b91ee1?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/d9e7d8b91ee1</guid>
            <category><![CDATA[cloud-careers]]></category>
            <category><![CDATA[data-infrastructure]]></category>
            <category><![CDATA[aws-certification]]></category>
            <category><![CDATA[cloud-infrastructure]]></category>
            <category><![CDATA[cloud-finops]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Sat, 13 Dec 2025 19:58:27 GMT</pubDate>
            <atom:updated>2026-04-10T02:17:00.987Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yC9uJLHvN4r-joYBmD-d3Q.jpeg" /><figcaption>Moi, en train de partager mon expérience lors de la Conférence EdTech UEMOA 2025, organisée par EtriLabs et la MasterCard Foundation</figcaption></figure><p>Pour ceux qui me suivent depuis plusieurs années sur les réseaux sociaux, je suis à la base, développeur informatique depuis bientôt 5 ans, ensuite suite à ma participation au programme intensif <a href="https://www.isheero.com/lancement-officiel-du-programme-africa-techup-tour-isheero/"><strong><em>Africa</em> <em>Tech Up Tour</em></strong></a><strong> en 2023</strong>, je me suis spécialisé dans le <strong>cloud</strong> <strong>Infrastructure engineering</strong>… Depuis lors j’ai progressé, appris, travaillé sur des projets, contribué à des initiatives open source, écrit des dizaines de contenus techniques sur <strong><em>l’écosystème cloud</em></strong> et <strong><em>FinOps</em></strong>.</p><p>Mais cette année a été particulière… — pas seulement à cause des “mois de disette” ou des centaines de milliers d’euros dilapidés dans des certifications AWS, Databricks, haha — mais particulière pour ce que j’ai fais, ce que j’ai accompli dans l’écosystème — sans fausse modestie ;⁠)</p><p>À l’aube d’une nouvelle année… c’est le moment donc de partager mon parcours professionnel, mes accomplissements, mes échecs, pour qu’ils inspirent et servent de mémoire pour la postérité.</p><h3><strong>A-/ Mes Certifications &amp; Accréditations obtenus</strong></h3><p>En novembre 2024, j’ai eu l’opportunité de démarrer un long programme en DevOps &amp; Cloud Engineering à <strong><em>l’Université Numérique Cheikh Hamidou Kane (UN-CHK) du Sénégal</em></strong>.</p><p>Programme m’ayant permis après un apprentissage et un suivi rigoureux, d’obtenir les certificats professionnels :</p><p><strong>✅ AWS Cloud Support Associate Professional Certificate (</strong><a href="https://www.coursera.org/account/accomplishments/specialization/42MWP7G85B7T"><strong>↪️ link</strong></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nGxTBrrB2nxJ7MecWYrGOA.jpeg" /><figcaption>AWS Cloud Support Associate Professional Certificate 🎓☁️⚙️</figcaption></figure><p><strong>✅ AWS Cloud Technology Consultant Professional Certificate ( in progress… / finally achieved 🫡 )</strong></p><p><a href="https://www.coursera.org/account/accomplishments/professional-cert/certificate/HX69X20A7FPG">(↪️link</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-cln3YYIvmWSqB52zxFedg.jpeg" /><figcaption>AWS Cloud Technology Consultant Professional Certified 🎓🎉</figcaption></figure><p>Pour continuer à affiner mes compétences techniques de l’écosystème cloud, j’ai aussi préparé, passé et réussi l’accréditation conjointe de Databricks et AWS ( Amazon Web Services ), il s’agit de :</p><p>✅<strong> Databricks Accredited AWS Platform Architect ( </strong><a href="https://credentials.databricks.com/d66ce504-2f31-4468-bd09-de9b6b8255ed"><strong>↪️ link</strong></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Dt3OLiqbwST1ma3YaB98yg.jpeg" /><figcaption>Databricks Accredited AWS Platform Architect Badge Officiel</figcaption></figure><p>Autres…</p><p>Mon année 2025 c’est aussi ma participation et la réussite à pleins d’autres programmes certifiants, en Kubernetes, FinOps, Gen AI.</p><p><strong>✅ FinOps + FOCUS ( Open Cost Usage Specification ) Certification de la célèbre FinOps Foundation</strong></p><p><strong>✅ Kubernetes and Cloud Native Certification de la Linux Foundation ( </strong><a href="https://www.credly.com/badges/e309adb5-2eb2-43b1-a788-0f141bb59cff/public_url"><strong>↪️link</strong></a></p><p><strong>✅ Alumni et Certifié du “AWS AI &amp; ML Scholars 2025”</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k9Soa1J308RL-2UynHamxw.jpeg" /><figcaption>AWS AI &amp; ML Scholars Alumni</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rFNWT-sioKJscvHBbL7zig.jpeg" /><figcaption>AWS Graduate 🎓</figcaption></figure><h3>B-/ Contributions Techniques &amp; Projets Cloud / FinOps</h3><p>Une chose étant d’apprendre, une autre est aussi d’appliquer, de pratiquer en situation réelle. C’est en ce sens que j’ai été <strong>Cloud FinOps Contributor</strong> au “<a href="https://kloo.me/PointFive-Hub"><strong><em>Cloud Efficiency Hu</em></strong><em>b</em></a>” de la célèbre startup israélo-américain <a href="https://www.pointfive.co/"><strong>PointFive</strong></a><strong>.</strong></p><p>Mes travaux là-bas ont porté sur <strong>l’optimisation des coûts</strong> pour la réduction des <strong>innéficacites sur le cloud de Microsoft Azure </strong>( oui oui, je fais pas que du AWS… 😁 )</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yhlgvFY1ZszT7DGvGY4L9Q.jpeg" /><figcaption>In efficient Use of Azure Pipelines by Orphéric ALLAGBE</figcaption></figure><p>Le résumé de mes travaux est consultable ici : ↪️<a href="https://kloo.me/PointFive-Hub">link</a></p><p>J’en ai également parlé dans ce post <a href="https://www.linkedin.com/posts/orph%C3%A9ric-allagbe-%F0%9F%91%A8%F0%9F%8F%BE%E2%80%8D%F0%9F%92%BB%E2%98%81%EF%B8%8F-1863961b6_my-first-finops-contribution-a-few-activity-7370802884848017408-Jy1Z?utm_source=share&amp;utm_medium=member_android&amp;rcm=ACoAADI_BnYBnk6dGjDx_LxpYy9iPNQerKDyaGA"><strong>LinkedIn ici</strong></a><strong>.</strong></p><p>Et comme, on ne dit jamais un sans deux, j’ai rejoins dans la foulée, la célèbre <a href="https://www.finops.org/"><strong>FinOps Foundation</strong></a> comme membre bénévole.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R9bZp79hNXVHvS0RnIxM2Q.jpeg" /><figcaption>FinOps Foundation Member Badge</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vGvUTnfbSRZ9aRPUpDHHuQ.jpeg" /><figcaption>Greetings from FinOps Foundation Board</figcaption></figure><h3>Mes contenus techniques sur le cloud infrastructure</h3><p>Étant Technical writer depuis plusieurs années déjà, j’ai eu à écrire des contenus techniques sur l’écosystème cloud, les architectures cloud natives, l’optimisation des coûts FinOps :</p><p>✅ <strong>AWS Community Builders</strong> : <a href="https://community.aws/@orphericloud">community.aws/@orphericloud</a></p><p><strong>🛑 2026 UPDATE: 🛑</strong></p><p><strong>Au regard de mes contributions et mes engagements dans la communauté cloud mondiale, j’ai été sélectionné par AWS, le 04 Mars 2026, en tant que AWS Community Builder, l’un des programmes les plus sélectifs et prestigieux d’AWS, dans le monde entier 😁🎉</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K7Re10iCdsY0v9YmuVrFoA.jpeg" /><figcaption>Mail de sélection reçu en tant que AWS Community Builder ☁️🎉</figcaption></figure><p>✅ Tous les articles sur mon <strong>blog Medium</strong> : <a href="https://kloo.me/Laravel-Technical-Writing">↪️ ici</a></p><p>L’un de mes meilleurs articles de cette année porte sur la <a href="https://medium.com/@Orpheric_Codeur/aws-wavelength-et-sonatel-orange-r%C3%A9volutionner-le-cloud-en-afrique-de-louest-avantages-dfcb685746a5"><strong>souveraineté cloud en Afrique : Cas du Sénégal</strong></a><strong>. </strong>J’y expose les tenants et aboutissants du déploiement d’<strong>AWS Wavelength</strong> au Sénégal.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iaERQpaKQaFCM82A1bUngg.jpeg" /><figcaption>AWS Community / @Orphericloud</figcaption></figure><h3>C-/ Events &amp; Programmes Internationaux</h3><h3><strong>⭐ AWS Cloud Club Community Program</strong></h3><p>Sélectionné par <strong>AWS</strong> comme <a href="https://www.credly.com/badges/e9cd0ae0-499f-4334-bb57-9c832faf8eca/linked_in_profile"><strong>Faculty Advisor</strong></a>, cette même année, j’ai sponsorisé et créé le tout premier Cloud Club Community dans mon pays le Bénin 🇧🇯</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_JDcL8J3LGS1ULsmI__QlQ.jpeg" /></figure><h3>⭐ NASA Space Apps Challenge 2025</h3><p><strong>Local Mentor</strong> des étudiants participants à la <strong><em>NASA Space Apps Challenge</em></strong>, une compétition de création d’applications répondant à des problématiques réelles dans le domaine de l’aéronautique, l’astronomie et les sciences spatiales.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KDuGnKoRh58i-ywC66Fhnw.jpeg" /><figcaption>Galactic Local Mentor Certification 🌠🚀🧑🏾‍🚀</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zN0NOdGpbenQsiHqA-0q0Q.jpeg" /><figcaption>Moi, (debout), en train de mentorer et accompagner les participants du Hackathon à Seme City 🎓</figcaption></figure><h3>⭐ AWS GameDay ☁️🎮🎯 Francophone 2025</h3><p>Sous la supervision d’AWS, et en tant que <strong>Co-Lead</strong> de la communauté <a href="https://www.meetup.com/aws-user-group-benin"><strong>AWS User Group Bénin</strong></a>, nous avions organisé dans les locaux d’<strong>Epitech Bénin</strong>, le tout premier <strong>GameDay</strong> <strong>AWS</strong>, une compétition où les équipes du Bénin ont affronté celles de la France, de la Belgique, et de la Suisse. Voir tout le récap de l’événement <a href="https://www.meetup.com/aws-user-group-benin/events/311116802/">ici 🔗</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nsYKzHhS7XveH81CN1hMbQ.jpeg" /><figcaption>First AWS GameDay at Bénin Republic 🇧🇯</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pgq_ll-hF9hTEjfmou4-Eg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*F5uGBc5O64JhK8PMWcQqCw.jpeg" /><figcaption>À Epitech Bénin, moi ( au milieu ), avec les participants au tout premier AWS GameDay 🎯🎮☁️</figcaption></figure><h3>Et cerise sur le gâteau… 🍰</h3><p>⭐ Je suis désormais <strong>AWS Academy Educator Certified.</strong> Grâce à cette certification, je suis désormais habilité et agréé pour enseigner les programmes universitaires du Cloud / Data infrastructure d’AWS dans les institutions académiques et universités technologiques.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*Npcc5osZYCWi8NYzFdkpUg.png" /></figure><p>Oui ce fut une année charnière entre apprentissages, certifications, projets, vie professionnelle, vie personnelle, avec son lot de disette ( j’ai fais 2–3 mois sans toucher à mon ordinateur… burn-out ), mais j’ai appris, j’ai encaissé chaque punch de la vie… J’ai échoué à des exams, j’ai été recalé à des programmes, j’ai perdu des admissions d’entrée… Je ne sais pas si mon année est réussie, car tant d’objectifs sont en suspens… mais je suis fier et extrêmement reconnaissant à la VIE, à Dieu, à ma famille, à mes amis proches et tout ceux qui m’ont aidé à un moment ou à un autre.</p><p><strong>JE VOUS AIME et JE VOUS EN SUIS RECONNAISSANT 🧡</strong></p><p>À l’année prochaine… 👏🏾</p><p>Sincerely,</p><p>Orphéric.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Xy8LKLUI6gZjAjKdTxO99g.jpeg" /><figcaption>Mon mantra. 🧡</figcaption></figure><h3>Alors qu’en est-il de 2026 ?</h3><h3>Je suis ouvert à de nouvelles opportunités professionnelles, de travail en lien avec l’une ou l’autre de mes compétences.</h3><p>💌 <strong>Mail Pro:</strong> <a href="/service/mailto:florentallagbe229@gmail.com">florentallagbe229@gmail.com</a></p><p>🌐 <strong>Site web:</strong> <a href="https://orphericallagbe.carrd.co/">orphericallagbe.carrd.co/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d9e7d8b91ee1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to deploy a whole Laravel app to AWS Lambda function ?]]></title>
            <link>https://medium.com/@Orpheric_Codeur/how-to-deploy-a-whole-laravel-app-to-aws-lambda-function-2c80499fb72d?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/2c80499fb72d</guid>
            <category><![CDATA[event-driven-architecture]]></category>
            <category><![CDATA[serverless]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[laravel-framework]]></category>
            <category><![CDATA[aws-lambda]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Wed, 23 Jul 2025 13:53:10 GMT</pubDate>
            <atom:updated>2025-07-23T13:53:10.672Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Dt6L79ccPFeJ0gkPN4oJmA.png" /></figure><p>Today, many tech companies are increasingly running their tech products on <strong>serverless infrastructures</strong>, given the advantages they offer in terms of scalability, high <strong>availability</strong>, and <strong>automatic scaling</strong>, not to mention the ability to not worry about how to provision or manage the server. It’s also worth noting that with its pay-per-use model, startups opting for <em>FinOps</em> and <em>cost optimization </em>have done everything they can to adopt it…</p><p><strong>Serverless functions</strong> sound exciting. But when I first heard about them, I had only one question: “All right. So we have these small files. Great. But how can I combine Laravel with functions? Should I ditch Laravel and write simple PHP code if I want a function? I won’t do that.”</p><p>Functions were originally designed to run very small services. However, in a larger application, this introduces a lot of complexity. Imagine a larger Laravel project you’ve been working on. Let’s say it has 100 API endpoints. Does that mean you have to refactor your monolithic application into 50+ services sitting on 50+ databases? Fortunately, no. This introduces problems you didn’t know existed. People won’t do it, but they want the convenience and infinite scalability of functions. It turns out we can easily deploy an entire Laravel application in a function.</p><p>This technical article is divided into two main parts. In the first, which will serve as a preliminary section, I will demonstrate in a basic way how to deploy a simple Python file on <strong>AWS</strong> <strong>Lambda </strong>using the <strong><em>serverless framework</em></strong>.</p><h4>Part A-)</h4><p><a href="#"><strong>Prerequisites</strong></a>: Have an AWS account (root or IAM user).</p><p><a href="#"><strong>Important Note:</strong></a> At the beginning of this article, I talked about <strong>serverless infrastructures</strong> and their importance. Throughout the rest, I’ll use the term “<strong>serverless framework”</strong>, which is a framework used in the ecosystem to develop and deploy <strong>AWS</strong> <strong>Lambda functions</strong> and other AWS services. Detailed <strong>documentation </strong><a href="http://serverless.com/">here</a>.</p><h4>SO DO NOT CONFUSE “SERVERLESS INFRASTRUCTURE/ARCHITECTURE” WITH “SERVERLESS FRAMEWORK”</h4><blockquote>LET’s GO !!!</blockquote><p>1- To install the required serverless framework, run the command:</p><pre>npm install -g serverless</pre><p>2- Log in with your IAM user or root user’s access keys <strong>( use your root user’s keys</strong> <strong>only for development purposes</strong> <strong>)</strong></p><pre>serverless config credentials --provider aws --key &lt;your-key&gt; --secret &lt;your-secret&gt;</pre><p>3- Now we’ll create the basic setup for our project to deploy. As explained at the beginning, we’re going to use Python for the testing phase. Project directory (my-service for example)</p><pre>serverless create --template aws-python3 --path my-service</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/261/1*3UlGWlAfL96ns5MlQkereA.jpeg" /><figcaption>Basics files created in the project directory</figcaption></figure><p>It also generates a <strong>serverless.yaml</strong> which is the main <strong>configuration</strong> for the serverless framework. serverless.yaml (or serverless.yml) is therefore the heart of the Serverless Framework configuration. It serves as a central <strong>declaration </strong>file to define, <strong>configure</strong> and <strong>deploy </strong>on the <strong>cloud</strong> (AWS, AZURE, GCP, etc.): It is <strong>INFRASTRUCTURE AS CODE</strong></p><pre># Welcome to Serverless!<br>#<br># This file is the main config file for your service.<br># It&#39;s very minimal at this point and uses default values.<br># You can always add more config options for more control.<br># We&#39;ve included some commented out config examples here.<br># Just uncomment any of them to get that config option.<br>#<br># For full config options, check the docs:<br>#    docs.serverless.com<br>#<br># Happy Coding!<br><br>service: my-service<br># app and org for use with dashboard.serverless.com<br>#app: your-app-name<br>#org: your-org-name<br><br># You can pin your service to only deploy with a specific Serverless version<br># Check out our docs for more details<br>frameworkVersion: &#39;3&#39;<br><br>provider:<br>  name: aws<br>  runtime: python3.9<br><br># you can overwrite defaults here<br>#  stage: dev<br>#  region: us-east-1<br><br>functions:<br>  hello:<br>    handler: handler.hello</pre><p>The <strong>most important </strong>thing about this file is the functions section. It defines a hello function with a handler of <strong>handler.hello</strong> . it refers to the <strong>handler.py</strong> file and the <strong>hello</strong> <strong>function </strong>inside it. And now we can deploy the <strong>function</strong>:</p><pre>serverless deploy</pre><p><a href="#"><strong>Note:</strong></a><strong> </strong>If you run “<em>serverless deploy</em>” it creates a <strong>.serverless</strong> folder. Inside that folder, you can see your project as a ZIP file, and the <a href="https://aws.amazon.com/cloudformation/"><strong>Cloudformation</strong></a> <strong>config </strong>generated by serverless.</p><p>Needless to say, before running this cmd, you need to go to the directory in your terminal ( <strong>cd </strong>my-service ) in my case.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/549/1*zyeGKy_QZGCTdHJlX_zcUQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/581/1*wHqnOCn3P3c6jogO1avNMw.jpeg" /></figure><p>Now let’s check if our <strong>serverless AWS</strong> <strong>Lambda function</strong> has been successfully deployed on AWS</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OJaVk5YBr-MLgaKM-DIILw.jpeg" /></figure><blockquote>Bingo !!</blockquote><p>You can invoke the <strong>function </strong>(<strong>hello</strong> in our case) using this command:</p><pre>serverless invoke -f hello</pre><p>It should the return this json:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/851/1*iRr1OL1UwzwLB59A2wyWaw.jpeg" /></figure><p>It’s all great but not very useful yet. Users won’t use the terminal to invoke our functions. And also, for just deploying a simple function we don’t really need a framework. AWS CLI can do that. DigitalOcean CLI can do that. The power of serverless comes from when you want to combine things. When you want to use more infrastructure than just a simple function.</p><p>So let’s expose the function via an <a href="https://aws.amazon.com/api-gateway/"><strong>API Gateway</strong></a>. An <strong>AWS API gateway </strong>is pretty similar to:</p><ul><li>An <strong>nginx </strong>reverse proxy we used with <strong>Docker</strong></li><li>An <strong>ingress</strong> we used with <strong>Kubernetes</strong></li></ul><p>It exposes an <strong>API</strong> to the outside world and then forwards the requests to the right services. In this case, a service is a <strong>Lambda function</strong>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5CmwSBAggVAAWmoGrAdJdg.png" /></figure><p>So <strong>serverless </strong>enables us to create an <strong>API gateway</strong> from the configuration. <strong>It’s basically infrastructure-as-code.</strong> Add this piece of code to your serverless.yaml file:</p><pre><br>functions:<br>  hello:<br>    handler: handler.hello<br>    events:<br>      - httpApi:<br>          path: /hello<br>          method: get</pre><p>I added an <strong>events section</strong> to the function. This means:</p><ul><li>There’s an <strong>event </strong>that will <strong>invoke </strong>the Lambda</li><li>The <strong>event</strong> is an <strong>HTTP request</strong> to the <em>/hello</em> endpoint</li></ul><p>If we now re-run “s<strong><em>erverless deploy</em></strong><em>”</em> we’ll get a message such as this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/694/1*FZ2XUQxerzZvv3PWYROFvg.jpeg" /></figure><p>There’s a URL for the function! Mine is this: <a href="https://6gjjfy241c.execute-api.us-east-1.amazonaws.com/hello">https://6gjjfy241c.execute-api.us-east-1.amazonaws.com/hello</a></p><p>In the <strong>AWS console</strong> you can see there’s a new <strong>API gateway</strong>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4ycuPCOl6V5G59PYGG1rqQ.jpeg" /></figure><p>In the <strong>API gateway</strong>, there’s a panel called <strong><em>Integration </em></strong>. Here you can see serverless successfully attached the lambda to the <strong>HTTP</strong> <strong>endpoint</strong>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VnH9HaJ3BenEjW1hbnZdiw.jpeg" /></figure><p>These are just the basics of <strong>serverless </strong>. It can create so many things for your application:</p><ul><li><strong>Websockets</strong></li><li><strong>Schedulers</strong></li><li><strong>Event streams</strong></li><li><strong>CloudWatch logging</strong></li><li><strong>S3 storage</strong></li></ul><p>It’s all configuration-driven which is a pretty good thing. Your repository contains the whole infrastructure that your application needs.</p><h4>CONGRATULATIONS IF YOU CAME THIS FAR, YOU HAVE JUST MADE A BIG STEP, ALMOST ALL OF IT !</h4><p>The title of this article was about deploying <strong>Laravel applications </strong>on <strong>AWS</strong>. So you’ve probably wondered why we didn’t deploy a simple <strong>PHP </strong>file instead of <strong>Python</strong> in our first part… If so, you’re a GENIUS! Haha…</p><p>The reason I used python in the part A) is that AWS Lambda doesn’t have a PHP runtime. It can be solved with 3rd party tools and AWS layers, however.</p><blockquote><strong>This is also the reason why I decided to write this article, when I encountered this problem on my cloud infrastructure.</strong></blockquote><h4>Part B-/</h4><p><a href="http://bref.sh/"><strong>Bref</strong></a><strong> </strong>is a composer package that provides <strong>php runtime</strong> on <strong>AWS</strong>, <strong>deployment </strong>tooling for PHP applications, and excellent <strong>Laravel</strong> <strong>integrations</strong>. It works together with <strong>serverless</strong> <strong>framework</strong>.</p><p>Let’s try it with a brand new Laravel installation:</p><pre>composer create-project laravel/laravel serverless-laravel</pre><p>Then:</p><pre>composer require bref/bref bref/laravel-bridge --update-with-dependencies</pre><p>After it’s installed, publish the serverless.yml config file:</p><pre>php artisan vendor:publish --tag=serverless-config</pre><p><strong>Important Note: If you have any problems installing the bref package (or incidentally bref/laravel-bridge), you can consult the following </strong><a href="http://bref.sh/docs/laravel/getting-started"><strong>documentation</strong></a><strong>.</strong></p><p>Let’s look at the generated config (<strong>serverless.yml</strong>) :</p><pre>service: laravel<br><br># Set your team ID if you are using Bref Cloud<br>#bref:<br>#    team: my-team-id<br><br>provider:<br>    name: aws<br>    # The AWS region in which to deploy (us-east-1 is the default)<br>    region: us-east-1<br>    # Environment variables<br>    environment:<br>        APP_ENV: production # Or use ${sls:stage} if you want the environment to match the stage<br>        SESSION_DRIVER: cookie # Change to database if you have set up a database<br><br>package:<br>    # Files and directories to exclude from deployment<br>    patterns:<br>        - &#39;!node_modules/**&#39;<br>        - &#39;!public/storage&#39;<br>        - &#39;!resources/assets/**&#39;<br>        - &#39;!resources/css/**&#39;<br>        - &#39;!resources/images/**&#39;<br>        - &#39;!resources/js/**&#39;<br>        - &#39;!storage/**&#39;<br>        - &#39;!tests/**&#39;<br>        - &#39;!database/*.sqlite&#39;<br>        # Exclude assets except for the manifest file<br>        - &#39;!public/build/**&#39;<br>        - &#39;public/build/manifest.json&#39;<br><br>functions:<br><br>    # This function runs the Laravel website/API<br>    web:<br>        handler: public/index.php<br>        runtime: php-82-fpm<br>        timeout: 28 # in seconds (API Gateway has a timeout of 29 seconds)<br>        events:<br>            - httpApi: &#39;*&#39;<br><br>    # This function lets us run artisan commands in Lambda<br>    artisan:<br>        handler: artisan<br>        runtime: php-82-console<br>        timeout: 720 # in seconds<br>        # Uncomment to also run the scheduler every minute<br>        #events:<br>        #    - schedule:<br>        #          rate: rate(1 minute)<br>        #          input: &#39;&quot;schedule:run&quot;&#39;<br><br>plugins:<br>    # We need to include the Bref plugin<br>    - ./vendor/bref/bref</pre><p>The <strong>provider</strong> is pretty similar to the earlier configs. It now contains an <strong>environment </strong>section. When you deploy with bref your .env file is going to be deployed as well, but here you can override values. package.patterns excludes files from the deployment.</p><p>The file defines two functions:</p><ul><li><strong>web</strong> is going to be the entire application. As you can see, it listens to every HTTP event. It means you get the frontend by visiting ‘<strong>/’</strong> and accessing the API by hitting an URL such as <strong>/api/users</strong> . The runtime <strong>php-82-fpm</strong> comes from the <strong>bref </strong>the plugin. This is what I talked about earlier. <strong>bref</strong> provides PHP runtimes on AWS.</li><li><strong>artisan </strong>is a function that runs on the <strong>php-82-console</strong> runtime which is a CLI PHP runtime and it enables us to run <strong>artisan </strong>commands. It’s similar to the official PHP CLI image on <strong>DockerHub</strong>.</li></ul><p>Before deploying with <strong>bref,</strong> you need to clear the <strong>config</strong> <strong>cache</strong>:</p><pre>php artisan config:clear</pre><p>Then just run:</p><pre>serverless deploy</pre><p>If you visit the URL given by command, you should see a fresh Laravel installation:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/864/1*5OlPrKwMseoVH6VqCpLeZQ.jpeg" /></figure><p>Let’s summarize what just happened: by running <strong>4 commands</strong> we deployed a new <strong>Laravel</strong> <strong>app</strong> to <strong>Amazon Lambda</strong>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2c80499fb72d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AWS Wavelength et Sonatel Orange : Révolutionner le Cloud en Afrique de l’Ouest — Avantages…]]></title>
            <link>https://medium.com/@Orpheric_Codeur/aws-wavelength-et-sonatel-orange-r%C3%A9volutionner-le-cloud-en-afrique-de-louest-avantages-dfcb685746a5?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/dfcb685746a5</guid>
            <category><![CDATA[souveraineté-numérique]]></category>
            <category><![CDATA[aws-africa]]></category>
            <category><![CDATA[aws-wavelength]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[innovation]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Thu, 15 May 2025 15:09:20 GMT</pubDate>
            <atom:updated>2026-10-13T14:36:07.928Z</atom:updated>
            <content:encoded><![CDATA[<h3>AWS Wavelength et Sonatel Orange: Révolutionner le Cloud en Afrique de l’Ouest -Avantages, Enjeux et Défis</h3><p>Comment le Sénégal vient de prendre un sérieux coup d’avance dans la révolution cloud en Afrique de l’ouest ?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/739/1*nDGc1fscTYKp2K7lJKRmvw.jpeg" /></figure><h4>🌍 Contexte : Une Collaboration Stratégique pour l’Afrique</h4><p>AWS (Amazon Web Services) et Sonatel Orange ont annoncé le déploiement d’une <a href="https://aws.amazon.com/about-aws/whats-new/2025/04/aws-wavelength-zone-dakar/"><strong>zone AWS Wavelength</strong></a> à Dakar, au Sénégal, en partenariat avec Orange Middle East &amp; Africa. Cette initiative, première en Afrique subsaharienne, marque une étape clé dans la démocratisation du cloud computing sur le continent, combinant infrastructure locale et services AWS de pointe.</p><p>Au lendemain de ce <a href="https://letechobservateur.sn/sonatel-et-aws-lancent-aws-wavelength-au-senegal-une-revolution-cloud-pour-lafrique-de-louest/"><strong>partenariat stratégique</strong></a>, en tant que professionnel architecte de solutions cloud, on est en droit de se demander quels sont les avantages et les défis qu’impliquent cela.</p><p>A travers cet article, j’userai de ma position d’Architecte et Consultant Cloud pour vous décortiquer les dits — et peut-être les non-dits — de ce deal.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/708/1*0Ny7LmNRIqCF6nhntK0anQ.jpeg" /></figure><h4>✅ Avantages Clés de ce Partenariat</h4><h4>1. Réduction Radicale de la Latence</h4><p><strong>AWS Wavelength</strong> intègre les services cloud d’AWS (EC2, EKS, EBS, etc.) directement dans les <strong><em>datacenters</em></strong> de <strong>Sonatel</strong>, connectés aux réseaux <strong>5G</strong> d’Orange. Cela permet une latence de <strong>quelques millisecondes</strong>, cruciale pour la création de solutions en :</p><ul><li><strong>Santé connectée</strong> : Télémédecine en temps réel et diagnostics par IA.</li><li><strong>Fintech</strong> : Transactions sécurisées et détection de fraude instantanée.</li><li><strong>Divertissement</strong>: Expériences sans délai pour les applications de <strong>gaming</strong>, et très approprié pour les applications de <strong>streaming HD.</strong></li></ul><h4>2. Souveraineté des Données Renforcée</h4><p>L’une des barrières à l’adoption du <strong>CLOUD HYBRIDE</strong> par les entreprises en Afrique, est la sécurité et la protection des données stockées localement. Ainsi, plusieurs entreprises privées — et mêmes publiques — préfèrent stockés leur données chez les fournisseurs cloud comme AWS, GCP, Azure, que de les confier aux datacenters locaux.</p><p>Mais grâce désormais à ce nouveau partenariat entre <strong>AWS et Sonatel Orange</strong>, les données sont hébergées localement dans le <strong>datacenter Tier III+</strong> de Sonatel à Dakar, répondant aux exigences réglementaires comme le futur <strong>RGPD africain</strong> et les lois sur la résidence des données. Un atout pour les secteurs régulés (banque, santé, éducation).</p><h4>3. Accès à une Infrastructure Cloud Globale</h4><p>Les entreprises africaines bénéficient des mêmes services AWS (EC2, EKS, ALB) qu’ailleurs, avec une intégration transparente aux régions AWS existantes. Cela facilite l’hybridation <strong>cloud/edge</strong> et l’innovation.</p><p>Jusque-là, seule la <a href="https://aws.amazon.com/about-aws/global-infrastructure/regions_az/">région AWS</a> <strong>af-south-1</strong> (Le Cap) est disponible en Afrique, limitant les régions AWS proches des utilisateurs finaux en Afrique de l’Ouest. Ainsi, — malgré la présence de la <a href="https://aws.amazon.com/about-aws/global-infrastructure/localzones/locations/">local Zone</a> <strong>af-south-1-los-1a</strong> — , AWS Wavelenght au Sénégal, vient faciliter plus nettement l’accès aux infrastructure cloud dans la sous-région en rapprochant désormais les utilisateurs finaux et les solutions numériques en Afrique subsaharienne.</p><h4>4. Soutien à l’Économie Numérique Locale</h4><ul><li><strong>Stratégie gouvernementale</strong> : Ce déploiement s’inscrit dans le <em>“New Technology Deal”</em> sénégalais (1,7 milliard $) pour accélérer la transformation digitale et positionner le pays comme une véritable plateforme tournante de l’économie numérique en Afrique de l’Ouest.</li><li><strong>Startups et PME</strong> : Accès à des outils cloud avancés (IA, IoT) sans investissement initial lourd, une bonne nouvelle pour l’optimisation des couts et le <strong><em>FinOps</em></strong>.</li></ul><h4>5. Optimisation des Coûts Opérationnels</h4><p><strong>Sonatel </strong>propose un <a href="https://espaceclientv2.orange.sn/digitaliser/offres-cloud/iaas/aws-wavelength"><strong>accompagnement</strong></a> complet : migration, optimisation des ressources, et support technique. Même Orange utilise cette infrastructure pour ses applications internes, réduisant ses coûts. Vous pouvez y souscrire par <a href="https://espaceclientv2.orange.sn/digitaliser/offres-cloud/iaas/aws-wavelength"><strong>ici</strong></a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/739/1*WwKbU0JByK4nv2TlUClAtg.jpeg" /><figcaption>Accord de Partenariat entre Sonatel Orange et Amazon Web Services (AWS) au GITEX 2024</figcaption></figure><h4>🛠️ Enjeux Techniques et Défis</h4><h4>1. Dépendance aux Infrastructures Physiques</h4><ul><li>La performance dépend de la fiabilité des réseaux 5G et de la fibre (corridor Lagos-Dakar). Toute rupture affecterait les services cloud.</li><li><strong>Couverture 5G limitée</strong> : Actuellement concentrée dans les zones urbaines (Dakar, Abidjan).</li></ul><h4>2. Conformité et Sécurité</h4><ul><li><strong>Risque extraterritorial</strong> : Le <a href="https://www.congress.gov/bill/107th-congress/house-bill/3162"><strong><em>USA PATRIOT Act</em></strong></a> ou le CLOUD Act américain pourraient théoriquement s&#39;appliquer à ce partenariat malgré la localisation des données au Sénégal.</li><li><strong>Alignement des normes</strong> : Harmoniser les standards de cybersécurité AWS avec les régulations locales (ANSSI Sénégal) et accélérer l’adoption du <strong>RGPD africain</strong> pour encadrer juridiquement les données.</li></ul><h4>3. Adoption par le Marché Local</h4><ul><li><strong>Manque d’expertise cloud</strong> : Pénurie de talents certifiés AWS en Afrique de l’Ouest. Des programmes de formation comme <a href="https://lecloudfacile.com/"><strong>LeCloudFacile</strong></a> au <strong>Sénégal</strong> et AWS re/Start offrent déjà des parcours de formations structurés et de qualité pour la prise en main de <strong>AWS Wavelength</strong>.</li><li><strong>Sensibilisation aux avantages du edge computing</strong> : Beaucoup d’entreprises ignorent encore son potentiel, il est temps que les DSI, les startups tech et les entreprises l’intègrent dans leur transformation digitale.</li></ul><h4>4. Modèle Économique Durable</h4><p>Face aux<strong> coûts d’accès </strong>les PME pourraient être freinées par les tarifs, malgré les économies à long terme. Des subventions publiques ou des partenariats public-privé sont nécessaires pour faciliter <a href="https://espaceclientv2.orange.sn/digitaliser/offres-cloud/iaas/aws-wavelength">l’accès</a> à cette technologie.</p><p>Les gouvernements pourraient s’appuyer sur cette infrastructure pour des projets nationaux (e-santé, éducation immersive).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*cgYVuWRvLbT-uKbzEDVU6A.png" /></figure><h4>📈 Cas d’Usage de solutions numériques</h4><p>Voici une liste de <strong>cas d’usage concrets</strong> pour <a href="https://docs.aws.amazon.com/wavelength/latest/developerguide/what-is-wavelength.html"><strong>AWS Wavelength</strong></a><strong> </strong>en Afrique, adaptés aux besoins locaux et aux défis technologiques du continent, avec des exemples réalistes et des ressources à l’appui :</p><h4>🏥 1. Santé connectée en zones rurales</h4><p><strong>Problématique</strong> : Manque de médecins spécialistes et infrastructures médicales dans les régions reculées.<br><strong>Solution avec </strong><a href="https://docs.aws.amazon.com/wavelength/latest/developerguide/what-is-wavelength.html"><strong>Wavelength</strong> </a>:</p><ul><li><strong>Télémédecine en temps réel</strong> : Consultation vidéo HD sans latence entre médecins urbains et patients ruraux, via des kiosques de santé équipés de capteurs IoT (ex. : ECG, pression artérielle).</li><li><strong>Diagnostics IA</strong> : Analyse instantanée de radios ou IRM par des modèles d’IA hébergés en edge (ex. : détection de tuberculose ou de paludisme via Amazon SageMaker).</li></ul><p><strong>Exemple</strong> : Le projet <a href="https://www.babyl.rw/"><em>Babyl Health</em></a> au Rwanda, combiné à AWS Wavelength, pourrait étendre ses services à des zones sans fibre optique.</p><h4>📱 2. Fintech et inclusion financière</h4><p><strong>Problématique</strong> : Latence des transactions mobile money entraînant des fraudes ou des abandonnements.<br><strong>Solution avec </strong><a href="https://docs.aws.amazon.com/wavelength/latest/developerguide/what-is-wavelength.html"><strong>Wavelength</strong> </a>:</p><ul><li><strong>Paiements instantanés</strong> : Traitement des transactions en &lt;10 ms, même lors des pics (ex. : paiements de salaires ou transferts transfrontaliers UEMOA).</li><li><strong>Détection de fraude en temps réel</strong> : Analyse comportementale via AWS <a href="https://aws.amazon.com/fr/lambda/edge/">Lambda@Edge</a> pour bloquer les transactions suspectes (ex. : fraude par SIM swap).</li></ul><p><strong>Exemple</strong> : <em>Wave Mobile Money</em> (Sénégal) ou <em>M-Pesa</em> (Kenya) pourraient réduire les coûts d’infrastructure tout en améliorant la sécurité.<br>🔗 <a href="https://aws.amazon.com/financial-services/">AWS Financial Services</a></p><h4>🌾 3. Agriculture intelligente</h4><p><strong>Problématique</strong> : Pertes post-récolte dues à des chaînes logistiques inefficaces.<br><strong>Solution avec </strong><a href="https://docs.aws.amazon.com/wavelength/latest/developerguide/what-is-wavelength.html"><strong>Wavelength</strong> </a>:</p><ul><li><strong>Capteurs IoT low-cost</strong> : Surveillance en temps réel de l’humidité des sols ou de la température des entrepôts, avec traitement des données en edge (ex. : déclenchement automatique d’irrigation via AWS IoT Core).</li><li><strong>Marchés agricoles décentralisés</strong> : Plateformes de vente en direct avec réalité augmentée (ex. : visualisation 3D du bétail pour les acheteurs étrangers).</li></ul><p><strong>Exemple</strong> : La start-up nigériane <em>Farmcrowdy</em> pourrait optimiser sa logistique avec des données traitées localement.</p><h4>🎮 4. Gaming et divertissement local</h4><p><strong>Problématique</strong> : Accès limité aux serveurs de jeux cloud (ex. : GeForce Now) à cause de la latence élevée.<br><strong>Solution avec </strong><a href="https://docs.aws.amazon.com/wavelength/latest/developerguide/what-is-wavelength.html"><strong>Wavelength</strong> </a>:</p><ul><li><strong>Cloud gaming low-cost</strong> : Jeux streamés depuis Dakar avec latence &lt;20 ms, accessibles via smartphones 5G (idéal pour des studios africains comme <em>Kiro’o Games</em> au Cameroun).</li><li><strong>Streaming de contenus locaux</strong> : Plateformes comme <em>Showmax Africa</em> pourraient héberger des serveurs vidéo en edge, réduisant les coûts de bande passante.</li></ul><p><strong>Exemple</strong> : <em>Carry1st</em> (Afrique du Sud), éditeur de jeux mobiles, pourrait lancer des tournois e-sport continentaux.<br>🔗 <a href="https://aws.amazon.com/gaming/">AWS for Gaming</a></p><h4>🚗 5. Mobilité urbaine et transports / Smart Cities</h4><p><strong>Problématique</strong> : Embouteillages chroniques dans les mégalopoles comme Dakar ou Lagos.<br><strong>Solution avec </strong><a href="https://docs.aws.amazon.com/wavelength/latest/developerguide/what-is-wavelength.html"><strong>Wavelength</strong> </a>:</p><ul><li><strong>Feux de circulation intelligents</strong> : Analyse vidéo en temps réel via Amazon Rekognition pour adapter les durées de feux tricolores.</li><li><strong>Bus connectés</strong> : Maintenance prédictive des véhicules grâce à des capteurs traitant les données en edge (ex. : détection de panne moteur avant qu’elle ne survienne).</li></ul><p><strong>Exemple</strong> : <em>GoMetro</em> (Afrique du Sud) pourrait optimiser ses solutions de gestion de flottes.<br>🔗 <a href="https://aws.amazon.com/automotive/">AWS Automotive</a></p><h4>🌍 6. Éducation immersive</h4><p><strong>Problématique</strong> : Accès inégal à des formations techniques de qualité.<br><strong>Solution avec </strong><a href="https://docs.aws.amazon.com/wavelength/latest/developerguide/what-is-wavelength.html"><strong>Wavelength</strong></a> :</p><ul><li><strong>Classes virtuelles en AR/VR</strong> : Élèves équipés de casques légers accèdent à des laboratoires virtuels (ex. : simulations de circuits électroniques ou de chirurgie).</li><li><strong>Tutoriels hors ligne</strong> : Stockage local de contenus éducatifs sur edge servers pour les zones rurales mal connectées.</li></ul><p><strong>Exemple</strong> : L’initiative <em>African Leadership University</em> pourrait déployer des formations en génie logiciel avec AWS.</p><p><strong>Et vous, quel cas d’usage prioritaire devraient développer AWS et Sonatel en Afrique ?</strong> Partagez avec moi votre avis en commentaire ! 👇</p><h4>💡 <a href="#">Conclusion</a>: Un Jalon pour la Souveraineté Numérique Africaine</h4><p>Ce partenariat A<strong>WS-Sonatel</strong> illustre à la fois le potentiel et les écueils du cloud en Afrique. Si les avantages (latence, souveraineté) sont tangibles, les défis (sécurité, compétences) exigent une réponse collective. En combinant innovation technologique et renforcement des capacités locales, l’Afrique pourrait écrire un nouveau chapitre de son indépendance numérique.</p><blockquote>Ecrit par <strong><em>Orphéric Florent</em></strong>,</blockquote><blockquote><strong>AWS Cloud Technology Consultant Certified.<br>AWS Cloud Support Associate Certified</strong>.</blockquote><blockquote>🔗 Contactez-moi via mon <a href="https://orphericallagbe.carrd.co/"><strong>site web</strong></a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dfcb685746a5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What are the common network troubleshooting steps for Amazon VPC ?]]></title>
            <link>https://medium.com/@Orpheric_Codeur/what-are-the-common-network-troubleshooting-steps-for-amazon-vpc-3ec122dbf185?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/3ec122dbf185</guid>
            <category><![CDATA[coud-security]]></category>
            <category><![CDATA[ec2]]></category>
            <category><![CDATA[amazon-web-services]]></category>
            <category><![CDATA[ec2-instance]]></category>
            <category><![CDATA[amazon-vpc]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Sat, 15 Mar 2025 15:50:14 GMT</pubDate>
            <atom:updated>2025-03-15T15:50:14.049Z</atom:updated>
            <content:encoded><![CDATA[<h3>What are the common network troubleshooting steps for Amazon VPC ?</h3><figure><img alt="Overview of IBM MQ architecture on AWS" src="https://cdn-images-1.medium.com/max/1024/1*LwubyR8qFCo6orIJ-dR94A.png" /><figcaption>Overview of IBM MQ architecture on AWS</figcaption></figure><p><a href="https://docs.aws.amazon.com/vpc/latest/userguide/what-is-amazon-vpc.html"><strong>Amazon VPC</strong></a><strong> (Virtual Private Cloud)</strong> is a service that allows you to create an isolated virtual network in the AWS cloud. Here are some key points about its role in designing a cloud architecture on AWS:</p><ul><li><strong>Isolation</strong>: VPC allows you to create an isolated network environment where you can deploy your AWS resources, such as EC2 instances, databases, etc.</li><li><strong>Network Control:</strong> You can configure subnets, route tables, and internet gateways to control inbound and outbound traffic to your resources.</li><li><strong>Security: VPC</strong> offers security features, such as security groups and access control lists (<strong>NACLs</strong>), to protect your resources.</li><li><strong>Connectivity</strong>: You can establish secure connections between your VPC and your on-premises networks via VPNs or Direct Connect.</li><li><strong>Scalability</strong>: VPC allows you to design scalable architectures by easily adding resources as needed.</li></ul><p>In summary, Amazon <strong>VPC </strong>is essential for building a secure and flexible cloud infrastructure on AWS.</p><p><strong>Context: </strong>You have a <strong>web application</strong> or <strong>SaaS </strong>running on an Amazon <strong>EC2 instance </strong>located in a public subnet of an <strong>Amazon VPC</strong>. There are several network configurations that determine whether your instance is accessible to the internet.</p><p>Below, we’ll list the configurations you should check if you ever have a public <strong>EC2</strong> <strong>instance </strong>with a web application that isn’t loading as expected.</p><blockquote>Let’s Go !!!</blockquote><p><strong>1. Internet Gateway</strong></p><p>Ensure an Internet Gateway (<strong>IGW</strong>) is attached to your <strong>VPC</strong>. Without an Internet Gateway, no traffic will be allowed to enter or exit the VPC.</p><p>With the Internet Gateway, the EC2 instance can send and receive internet traffic. Users can access your web application using the instance’s public IP address.</p><p>Without the Internet Gateway, the EC2 instance would be unable to communicate with the internet, making the web application inaccessible to external users. The Internet Gateway is therefore crucial for enabling access to your resources within the VPC.</p><p><strong>2. Route Tables</strong></p><p>Check the route table associated with your EC2 instance’s subnet. Ensure there is a route with a destination of <strong>0.0.0.0/0</strong> <strong>pointing </strong>to the <strong>Internet</strong> <strong>Gateway</strong>. This route allows outbound traffic to the internet.</p><p>By configuring the routing table, the <strong>EC2 instance</strong> can send requests to the internet (for example, to download updates or access <strong>external APIs</strong>) and receive responses.</p><p>Without the <strong>0.0.0.0/0</strong> <strong>route </strong>pointing to the internet gateway in the routing table, the <strong>EC2 instance</strong> would be unable to access the internet, thus limiting its functionality. The routing table is therefore essential for enabling outbound traffic to the internet.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/521/1*ZTk4IEhMv4OsMKujExB17A.png" /></figure><p><strong>3. Security Groups</strong></p><p>Review the security group settings attached to your <strong>EC2 instance</strong>. Verify that there are inbound rules allowing <strong>HTTP</strong> (<strong>port 80</strong>) and/or <strong>HTTPS </strong>(<strong>port 443</strong>) traffic from the internet (<strong>0.0.0.0/0</strong>). Also verify that the egress rules allow traffic to leave the instance.</p><p>Without the proper rules in security groups, the <strong>EC2 instance </strong>would be unable to receive <strong>HTTP</strong> or <strong>HTTPS</strong> <strong>traffic</strong>, rendering the web application inaccessible. <strong>Security groups </strong>are therefore essential for controlling access and securing your resources.</p><p><strong>4. Network Access Control List</strong></p><p>Check the <strong>Network Access Control Lists</strong> (<strong>NACLs</strong>) associated with your subnet. Ensure they allow inbound and outbound traffic for <strong>HTTP</strong> and <strong>HTTPS</strong>. Unlike security groups, <strong>NACLs</strong> are <strong>stateless</strong>, so you must explicitly allow inbound and outbound rules.</p><p><strong>5. Public IP Address</strong></p><p>Ensure your <strong>EC2 instance</strong> has been assigned a public <strong>IP address</strong>. You can verify this in the EC2 console under the instance details. If it doesn’t have a public IP address, reboot the instance and ensure the configuration for assigning a public IP address is correct.</p><p><strong>6. HTTP vs. HTTPS</strong></p><p>Confirm that your application is accessible via the correct <strong>protocol</strong>. If your application is configured for HTTPS, ensure the <strong>SSL/TLS</strong> certificates are correctly installed and configured. Also check if your web browser is trying to connect via the wrong protocol (HTTP instead of HTTPS or vice versa). For this course, the application runs over HTTP, so check that your browser isn’t trying to connect via HTTPS. You can do this by selecting the browser’s address bar and making sure the address begins with http and not https.</p><p><strong>7. User Data Script</strong></p><p>If your instance uses a user data script to configure the application at launch, verify that the script ran successfully. Check the instance logs (<strong>/var/log/cloud-init.log</strong> or <strong>/var/log/cloud-init-output.log</strong>) for any errors that may have occurred during the execution of the user data script.</p><p><strong>8. Permissions</strong></p><p>Check the permissions and roles attached to your EC2 instance. Ensure the instance has the necessary <strong>IAM roles</strong> and <strong>policies </strong>to access all required AWS services, such as S3, DynamoDB, or RDS.</p><blockquote><strong>Note: </strong>I also wrote a useful <a href="https://medium.com/@Orpheric_Codeur/security-best-practices-for-your-aws-accounts-using-identity-and-access-management-iam-34f48e46bb0f"><strong>article</strong></a><strong> </strong>on.</blockquote><p><a href="https://medium.com/@Orpheric_Codeur/security-best-practices-for-your-aws-accounts-using-identity-and-access-management-iam-34f48e46bb0f">Security Best Practices for Your AWS Accounts Using Identity and Access Management (IAM)</a></p><p><strong>9. Personal Network Permissions</strong></p><p>Ensure your personal or corporate network does not have restrictions blocking access to your EC2 instance’s public IP address. Some networks may have firewalls or proxy settings that could block outbound traffic to certain IP ranges or ports.</p><p><strong>10. Application</strong></p><p>Ensure your application code is deployed and running correctly. Check the application logs to diagnose any runtime errors. Also, ensure your web server (e.g., Apache, Nginx) is installed and running.</p><p>In summary, errors on your networks running on Amazon VPC are common and more or less related to your architecture, so it is necessary to keep in mind these steps and possible solutions to help you in the future when you need them.</p><h4><a href="https://medium.com/@Orpheric_Codeur/security-best-practices-for-your-aws-accounts-using-identity-and-access-management-iam-34f48e46bb0f#">Epilogue</a>: You are a tech company, a startup or an individual actively needing a Cloud DevOps Engineer, to use and effectively integrate the cloud into your processes, let’s discuss it here via email: <a href="/service/mailto:florentallagbe229@gmail.com">florentallagbe229@gmail.com</a> or <a href="https://linkedin.com/in/orpheric-allagbe-1863961b6">on my website</a>.</h4><h4>I also train IT teams in DevOps / Cloud best practices.</h4><h4>I am Orphéric Florent,</h4><h4>AWS Cloud Technology Consultant Certified.<br>AWS Cloud Support Associate Certified.</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3ec122dbf185" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How can SMEs, companies and startups use “cloud computing” technology to initiate their development…]]></title>
            <link>https://medium.com/@Orpheric_Codeur/how-can-smes-companies-and-startups-use-cloud-computing-technology-to-initiate-their-development-1d1699bdaca6?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/1d1699bdaca6</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[cloud-engineering]]></category>
            <category><![CDATA[aws-cloud]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Sun, 09 Mar 2025 12:46:09 GMT</pubDate>
            <atom:updated>2025-03-09T13:12:09.223Z</atom:updated>
            <content:encoded><![CDATA[<h3>How can SMEs, companies and startups use “cloud computing” technology to initiate their development and achieve growth, [in the AI ​​era]?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/626/1*E7Tt5TPp8MzfgRPVyIxLBg.jpeg" /><figcaption>Source: FreePik</figcaption></figure><p>Since the end of 2022, when generative AI was democratized thanks to ChatGPT, AI has become a real fad and has gradually and significantly changed our habits in terms of production and design. Almost all companies today integrate — or seek to integrate — AI in one way or another into their process.</p><p>But during this massive rush towards AI, one fact catches my attention: how many companies are integrating cloud computing technology into their processes like the thousands of AI tools used all day long?</p><p>According to a <a href="https://www.agenceecofin.com/telecom/0107-119900-en-afrique-subsaharienne-27-des-dirigeants-d-entreprise-ont-adopte-l-ia-generative-rapport"><strong>study </strong></a>by the <strong>ecofin agency </strong>in 2024, in sub-Saharan Africa, a study indicates that 12% of business leaders have adopted cloud computing across their operations, while 38% have integrated it into a large part of their activities. Furthermore, 32% of companies have adopted the cloud in some of their operations, and 19% do not yet use this type of solution.</p><p>Given the statistics, it is clear to see how things are going “badly and better” despite the growing number of companies adopting cloud computing, Aha!</p><p>This article explores how to leverage the cloud to unlock business growth potential, with a focus on key concepts like <strong>virtualization</strong>, <strong>AWS services</strong>, and <strong>virtual private networks</strong> (VPNs).</p><blockquote>Enough chatter… Let’s get to the heart of the matter.</blockquote><p>In Africa, cloud computing has become a catalyst for digital transformation, providing SMEs and startups with tools that were once reserved for large enterprises. With providers like <a href="https://aws.amazon.com/"><strong>AWS</strong></a>, this technology helps overcome infrastructure, cost and security challenges, while accelerating innovation and value creation.</p><p><strong>A- Cloud computing for dummies</strong></p><p>When we talk about the cloud, for many people, it’s “cloud storage”. Instant confession: I thought the same thing about ten years ago. Haha!</p><p>The good news is that you weren’t really wrong if you thought that way. But the truth is that “cloud storage” as you call it, is also part of the services that make up Cloud technology. <a href="#"><strong>Spoiler</strong></a>: We have thousands of services in the cloud. Even AI is hosted in the cloud and uses data centralized in the cloud. Wow!</p><p>About two to three decades ago, for companies to create their own software and store their user data internally, they had to invest, buy and install servers, configure networks, and many others. So as the company grew, it had to provide more expensive hardware resources. Ultimately, scalability is an issue.</p><p>This is where cloud computing comes into play as the ultimate solution to the costly expenses of the operational departments (<strong>ops</strong>) of companies.</p><h4>The cloud therefore allows remote access, via the internet, to resources (storage, virtualization, computing, network, AI, IoT, etc.) that a local physical infrastructure would offer.</h4><p>This is therefore somewhat the case for most companies today that use remote server resources for their products and services without having physical servers internally on their premises.</p><p>Is that all? Yes. The cloud is that simple. Haha! Too simple, right?!</p><p><strong>B- On-premises Infrastructure VS Cloud Infrastructure</strong></p><p><strong>On-premises infrastructure</strong> refers to servers, hardware, and software that are physically installed on a company’s premises. In this case, the company has full control over the infrastructure, but its maintenance requires high costs.</p><p><strong>Cloud infrastructure</strong>, on the other hand, uses computing resources provided by cloud service providers via the internet such as (<a href="https://aws.amazon.com/"><strong>AWS</strong></a>, <a href="https://azure.microsoft.com/en-us/"><strong>Microsoft Azure</strong></a>, <a href="https://cloud.google.com/"><strong>Google Cloud</strong></a>). Here, control of the infrastructure is shared with the cloud service provider that manages the underlying infrastructure.</p><p>Here, the costs are usually based on a pay-as-you-go model, which helps reduce unnecessary expenses and allows you to pay only for what you use.</p><p><strong>C- Practical Cases: How can companies leverage the Amazon Web Services (AWS) cloud to solve complex problems?</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*j48lSQU_uqCN6nqWTNeUEw.jpeg" /><figcaption>Image illustrating the Amazon Web Service (AWS) cloud. Generated by DALL-E</figcaption></figure><p><strong>Problem 1-): Secure Sharing of Internet Connection via a Cloud VPN (Virtual Private Network) for a Consulting Firm</strong></p><p>At a Morocco-based consulting firm with 200 employees working remotely across Africa, employees were accessing the internet through unsecured home networks, exposing client data (sensitive market research) to risk of leaks or interception.</p><h4>AWS Cloud Solution: VPN Architecture with Secure Internet</h4><p>The goal here is to tunnel all employee Internet traffic through the AWS cloud to:</p><ul><li>Encrypt data.</li><li>Apply security policies (DNS filtering, site blocking).</li><li>Centralize access logs.</li></ul><p><strong>Amazon Web Services (AWS) Cloud Services Used:</strong></p><ol><li><a href="http://aws.amazon.com/vpn/client-vpn"><strong>AWS Client VPN</strong></a>: To connect remote users to the <strong>VPC (Virtual Private Cloud).</strong></li><li><a href="https://aws.amazon.com/network-firewall/"><strong>AWS Network Firewall</strong></a>: To filter outbound traffic (application layer).</li><li><a href="https://aws.amazon.com/route53/resolver-dns-firewall/"><strong>Amazon Route 53 DNS Firewall</strong></a>: Blocks <strong>DNS </strong>queries to malicious domains and phishing and cryptomining sites</li><li><a href="https://aws.amazon.com/fr/cloudwatch/"><strong>Amazon CloudWatch</strong></a>: Monitoring connection <strong>logs </strong>and <strong>traffic</strong>.</li></ol><p><strong>Problem 2-) Securing bank transactions with a VPN Cloud (Fintech, Nigeria)</strong></p><p>A Nigerian fintech offering mobile payment services had to secure access to its sensitive databases (customer data, transaction historical) from remote agencies. A <strong>VPN on-site</strong> infrastructure (physical firewall) was expensive and complex to extend, it turns to a more optimal cloud solution.</p><p><strong>AWS solution:</strong></p><ul><li><a href="http://aws.amazon.com/vpn/client-vpn"><strong>AWS VPN customer</strong></a>: deployment of a fully managed virtual network, connecting agencies to an isolated VPC (Virtual Private Cloud).</li><li><a href="https://aws.amazon.com/fr/blogs/france/limportance-du-chiffrement-comment-aws-vous-aide-a-proteger-vos-donnees/"><strong>AES-256 encryption</strong></a><strong> via IPSEC</strong>, natively integrated.</li><li><strong>Access management via </strong><a href="https://aws.amazon.com/fr/iam/"><strong>IAM</strong> </a>(prohibition of outside Nigeria connections to prevent geolocated attacks).</li></ul><p><strong>Result :</strong></p><ul><li>Reduced costs of 50 % vs a on -site infrastructure solution.</li><li>Simplified <strong>PCI DSS</strong> compliance thanks to <strong>AWS certifications</strong>.</li></ul><p><strong>Problem 3-) Virtualization and scalability for an e-commerce platform (Ivory Coast)</strong></p><p>An Ivorian marketplace was experiencing x10 traffic spikes during the end-of-year sales, leading to local server crashes (response time / latency &gt; 10s).</p><p><strong>AWS Solution:</strong></p><ol><li><strong>Auto Scaling Group (</strong><a href="https://aws.amazon.com/fr/ec2/"><strong>Amazon EC2</strong></a><strong>)</strong>: Automatic addition of t4g.medium instances during peaks, based on <a href="https://aws.amazon.com/fr/cloudwatch/"><strong>CloudWatch </strong></a>metrics (CPU &gt; 70%).</li><li><strong>Containerization (</strong><a href="https://docs.aws.amazon.com/AmazonECS/latest/developerguide/AWS_Fargate.html"><strong>ECS Fargate</strong></a><strong>)</strong>: Migration of microservices to serverless containers, eliminating server management.</li><li><a href="https://aws.amazon.com/rds/aurora/"><strong>Amazon Aurora</strong></a>: MySQL-compatible database with Multi-AZ replication, guaranteeing latency &lt; 3 ms.</li></ol><p><strong>Result:</strong></p><ul><li>Management of 50,000 requests/minute during sales, without interruption.</li><li>Optimized costs thanks to the automatic shutdown of unused instances.</li></ul><p><strong>Problem 4-) Predictive Analytics for Agriculture (Kenya) — AI and Storage</strong></p><p>A farming cooperative wanted to predict crop yields by analyzing satellite images (10TB/month) and weather data. Local servers lacked computing power.</p><p><strong>AWS Solution:</strong></p><ul><li><strong>Storage</strong>: Image archiving in <a href="https://aws.amazon.com/s3/storage-classes/glacier/"><strong>Amazon S3 Glacier</strong></a> ($ 0.004/GB/month).</li><li><strong>Calculation</strong>: Use of <a href="https://aws.amazon.com/fr/ec2/"><strong>EC2 </strong></a><strong>P3</strong> (GPU NVIDIA V100) to cause a computer vision model via <a href="https://aws.amazon.com/sagemaker/"><strong>Sagemaker</strong></a>.</li><li><strong>Data pipeline:</strong> <a href="https://aws.amazon.com/glue/"><strong>AWS Grue </strong></a>to clean and prepare weather data (CSV/JSON format).</li></ul><p><strong>Result :</strong></p><ul><li>92 % precision on yield predictions, allowing harvesting optimization.</li><li>Reduction of storage costs by 70 % vs a on -site infrastructure solution.</li></ul><p><strong>Problem 5-) Telemedicine and video streaming (South Africa)-Network and calculation in the cloud</strong></p><p>A telemedicine startup had difficulty broadcasting HD video consultations in real time to rural areas, with high latency (&gt; 500 ms).</p><p><strong>AWS solution:</strong></p><ul><li><a href="https://aws.amazon.com/fr/ivs/"><strong>Amazon IVS </strong></a><strong>(Interactive Video Service)</strong>: Managed service for Low-Latency streaming (&lt; 3S), integrating CDN via <a href="https://aws.amazon.com/fr/cloudfront/"><strong>Cloudfront</strong></a>.</li><li><a href="https://aws.amazon.com/global-accelerator/"><strong>AWS Global Accelerator:</strong></a> Optimized traffic routing to the nearest AWS presence point (eg Johannesburg).</li><li><a href="https://aws.amazon.com/lambda/"><strong>AWS Lambda</strong></a>: Trigger of automated workflows (e.g. sending medical reports after each consultation).</li></ul><p><strong>Result :</strong></p><ul><li>Video broadcast in 1080p with latency &lt;200 ms, even in 3G.</li><li>30,000 consultations/month without investment in video equipment.</li></ul><blockquote>And one last for the end … Haha!</blockquote><p><strong>Problem 6-) Management of logistics fleets (Senegal)-IoT and virtualization</strong></p><p>A logistics company had to follow 500 trucks in real time, but its local servers cannot manage the volume of IoT data (GPS, temperature sensors).</p><p><strong>AWS solution:</strong></p><ol><li><a href="https://aws.amazon.com/iot-core/"><strong>AWS IoT Core</strong></a>: collection and processing of 10,000 messages/minute from the sensors via <a href="https://mqtt.org/"><strong>MQTT</strong></a>.</li><li><strong>Virtualization of vehicles</strong>: creation of “<a href="https://aws.amazon.com/fr/what-is/digital-twin/#:~:text=Les%20jumeaux%20num%C3%A9riques%20s&#39;appuient,%C3%A0%20jour%20en%20temps%20r%C3%A9el."><strong>digital twins</strong></a>” in <a href="https://aws.amazon.com/dynamodb/"><strong>Dynamodb</strong></a>, reflecting the real state of trucks (eg temperature, location).</li><li><a href="https://aws.amazon.com/kinesis/"><strong>Amazon Kinesis Data Analytics</strong></a>: Detection of anomalies (eg itinerary deviation) in real time.</li></ol><p><strong>Result :</strong></p><ul><li>20 % reduction in fuel consumption thanks to the optimization of roads.</li><li>Real -time alerts in the event of a break in the cold chain.</li></ul><p>From AI, streaming, IoT, calculation, storage and networks, we had seen by these cases of problem solving, how cloud computing technology through cloud suppliers and their services, can be a strategic tool for businesses and startups, just as generative AI is today.</p><p>During this article, I expressly focused on the services of Amazon (<strong>AWS</strong> <strong>Cloud</strong>). So let’s not forget the other suppliers including Google Cloud, Microsoft Azure and many others.</p><h4><a href="https://medium.com/@Orpheric_Codeur/security-best-practices-for-your-aws-accounts-using-identity-and-access-management-iam-34f48e46bb0f#">Epilogue</a>: You are a tech company, a startup or an individual actively needing a Cloud DevOps Engineer, to use and effectively integrate the cloud into your processes, let’s discuss it here via email: <a href="/service/mailto:florentallagbe229@gmail.com">florentallagbe229@gmail.com</a> or <a href="https://linkedin.com/in/orpheric-allagbe-1863961b6">on my website</a>.</h4><h4>I also train and educate IT teams in DevOps best practices.</h4><h4>I am Orphéric Florent,</h4><h4>AWS Cloud Technology Consultant Certified.<br>AWS Cloud Support Associate Certified.</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1d1699bdaca6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Security Best Practices for Your AWS Accounts Using Identity and Access Management (IAM)]]></title>
            <link>https://medium.com/@Orpheric_Codeur/security-best-practices-for-your-aws-accounts-using-identity-and-access-management-iam-34f48e46bb0f?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/34f48e46bb0f</guid>
            <category><![CDATA[aws-iam-policy]]></category>
            <category><![CDATA[cloud-security]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[aws-iam-best-practices]]></category>
            <category><![CDATA[iam-roles]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Thu, 06 Mar 2025 14:29:18 GMT</pubDate>
            <atom:updated>2025-03-06T14:56:47.105Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/876/1*iUgXjDVU47IuV0F6O-AQJA.png" /></figure><p>For all cloud solution architects on <strong>Amazon Web Services</strong> (AWS), the frequent use of <a href="https://aws.amazon.com/iam/"><strong>AWS IAM</strong></a> (Identity Access Management) has become essential because of its role in the management and security of an AWS account and its resources.</p><h3>What’s IAM ?</h3><p><a href="https://aws.amazon.com/iam/"><strong>IAM </strong></a>is the ultimate access management service for <strong>AWS</strong>. It allows you to manage login credentials and permissions for your <strong>AWS account </strong>and can also manage credentials used to sign <strong>API </strong>calls made to AWS services.</p><p><a href="https://aws.amazon.com/iam/"><strong>IAM</strong></a><strong> </strong>allows you to create users and each person who needs access to your <strong>AWS</strong> <strong>acount</strong> has their own unique IAM user.</p><blockquote><strong>Important Note: </strong>EVERYTHING IN AWS ​​IS AN API CALL</blockquote><p>Each action that is carried out in AWS is an <strong>authenticated</strong> and <strong>authorized</strong> <strong>API </strong>call. We can therefore make API calls to services and resources.</p><p>Then, all <strong>API calls</strong> in AWS must be signed and authenticated to be authorized, whether or not the resources are in the same <strong>AWS</strong> <strong>account</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/360/1*skDJXa_hwZR4Yb5-fdv16A.png" /></figure><h3>AWS IAM Best Practices #1</h3><p>Organize users into groups and assign permissions to groups rather than individual users.<br>This makes it easier to manage when people change roles or multiple users need permissions applied or revoked.</p><h4>Application context:</h4><p>In a tech company, you have three teams: <strong>Devs</strong>, <strong>DevOps</strong>, <strong>Accounting</strong>.<br>We can create three IAM groups: Developers have “<strong>read-only</strong>” access to an S3 bucket containing the source code, DevOps have <strong>EC2FullAccess</strong> and <strong>Lambda </strong>Execution <strong>permissions</strong> to manage the infrastructure and Accounting have access to <strong>AWS</strong> <strong>Billing</strong> to track costs and expenses.</p><h4>Usage scenario:</h4><ul><li>When a new employee joins the DevOps team, you add him to the “<strong>DevOps</strong>” <strong>IAM</strong> <strong>group </strong>and he immediately inherits the rights necessary to use the resources and services of the <strong>AWS account</strong>.</li><li>When a developer changes teams to become DevOps: you remove him from the “<strong>Developers</strong>” <strong>IAM group</strong> and add him to the “DevOps” group. His old permissions are automatically revoked.</li><li>When the company adds a new <strong>security policy </strong>for DevOps (e.g. restriction to an AWS region), you only need to modify a single IAM policy attached to the <strong>DevOps group</strong>, and all its members are impacted -positively-.</li></ul><p>This <strong>best</strong> <strong>practice </strong>is a huge time saver during security <strong>audits </strong>and helps avoid individual <strong>permission </strong>errors.</p><h3>AWS IAM Best Practices #2</h3><ul><li>When you create your AWS account, you need to set up <strong>MFA </strong>for the <strong>root</strong> <strong>user</strong>.</li><li>Then create an <strong>IAM user</strong> with admin permissions.</li><li>Log out of the root user and log in to the IAM user you just created. From there you can use this admin user to create the rest of the IAM <strong>groups</strong>, <strong>roles</strong>, <strong>users</strong>, and <strong>policies</strong>.</li><li>Consider removing the root user’s access keys or don’t share these root accesses.</li></ul><h4>Usage Scenario:</h4><p><strong>TechStartup Inc.</strong> is migrating its application to AWS. Here’s how it applies your best practices to secure its environment:</p><p><strong>Step 1:</strong> Set up MFA for the root account<br><strong>TechStartup’s Cloud Engineer</strong> creates the AWS account. Immediately enables MFA for the root account.<br>Why? If root credentials are compromised (e.g. password leak), MFA blocks unauthorized access.</p><p><strong>Step 2: </strong>Create an IAM Admin User<br>The <strong>Cloud Engineer </strong>creates an <strong>IAM</strong> <strong>AdminUser</strong>, attaches the <strong>AdministratorAccess </strong>managed policy to it, and sets up MFA for this user.</p><pre># Create the AdminUser user<br>aws iam create-user --user-name AdminUser<br><br># Attach the AdministratorAccess policy<br>aws iam attach-user-policy --user-name AdminUser --policy-arn arn:aws:iam::aws:policy/AdminisratorAccess</pre><p><strong>Step 3:</strong> Switch to AdminUser<br>The AWS Cloud Engineer logs out of the root account and logs back in with AdminUser. From now on, all infrastructure (IAM groups, roles, etc.) is managed through this user, without ever reusing the root account except for critical cases (e.g. closing the account).</p><p><strong>Step 4:</strong> Create Secure Groups/Users<br><strong>Actions:</strong><br>-Create a Developers group with read-only access to EC2 and S3.<br>-Create a Finance group with access to AWS Billing.<br>-Add users to groups based on their roles.</p><p><strong>Sample Developers</strong> <strong>policy</strong>:</p><pre>{<br>    &quot;Version&quot;: &quot;2012-10-17&quot;,<br>    &quot;Statement&quot;: [{<br>        &quot;Effect&quot;: &quot;Allow&quot;,<br>        &quot;Action&quot;: [&quot;ec2:Describe*&quot;, &quot;s3:Get*&quot;],<br>        &quot;Resource&quot;: &quot;*&quot;<br>    }]<br>}</pre><p><strong>Step 5:</strong> Delete Root Account Access Keys<br><strong>TechStartup </strong>checks via the <strong>IAM Dashboard</strong> that no Access Keys are active for the root account. If any keys exist, they are <strong>deleted</strong>.<br><strong>Why? Root keys allow unlimited access and are a prime target for attacks.</strong></p><pre># List root keys (run as root)<br>aws iam list-access-keys<br><br># Delete a root key<br>aws iam delete-access-key --access-key-id AKIAEXAMPLEKEY</pre><h4>IAM Policies and IAM Roles:</h4><p>IAM Policies define the permissions for the actions of users, groups and roles in an AWS account.</p><p>Most IAM policies are stored in JSON format.</p><p>IAM roles, on the other hand, have an identity that can be assumed by someone or something that needs temporary access (usually between 15mn — 36H) to <strong>AWS</strong> <strong>credentials</strong>, unlike IAM users who have long-term credentials.</p><p>It is therefore easier to manage roles than users.</p><h4>Additional resources:</h4><ul><li><a href="https://aws.amazon.com/iam/">Access Management- AWS Identity and Access Management (IAM) - AWS</a></li><li><a href="https://aws.amazon.com/blogs/security/how-to-create-and-manage-users-within-aws-sso/">How to create and manage users within AWS IAM Identity Center | Amazon Web Services</a></li><li><a href="https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html">Security best practices in IAM</a></li></ul><h4><a href="#">Epilogue</a>: You are a tech company, a startup or an individual actively needing a Cloud DevOps Engineer, to use and effectively integrate the cloud into your processes, let’s discuss it here via email: <a href="/service/mailto:florentallagbe229@gmail.com">florentallagbe229@gmail.com</a> or <a href="https://linkedin.com/in/orpheric-allagbe-1863961b6">on my website</a>.</h4><h4>I also train IT teams in DevOps best practices.</h4><h4>I am Orphéric Florent,</h4><h4>AWS Cloud Technology Consultant Certified.<br>AWS Cloud Support Associate Certified.</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=34f48e46bb0f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Comment les PMEs, entreprises et startups peuvent utiliser la technologie du “cloud computing” pour…]]></title>
            <link>https://medium.com/@Orpheric_Codeur/comment-les-pmes-entreprises-et-startups-peuvent-utiliser-la-technologie-du-cloud-computing-pour-23cd51799cd3?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/23cd51799cd3</guid>
            <category><![CDATA[aws-cloud]]></category>
            <category><![CDATA[cloud-engineering]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Sat, 15 Feb 2025 18:09:44 GMT</pubDate>
            <atom:updated>2025-03-09T13:08:06.589Z</atom:updated>
            <content:encoded><![CDATA[<h3>Comment les PMEs, entreprises et startups peuvent utiliser la technologie du “cloud computing” pour amorcer leur développement et atteindre la croissance, [ à l’ère de l’IA ] ?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UokosZh-2ha_F6u1FZXlxQ.jpeg" /><figcaption>image illustrant le cloud computing. Générée par DALL-E</figcaption></figure><p>Depuis fin 2022, ou l’IA générative a été démocratisé grâce à ChatGPT, l’IA est devenu un vrai effet de mode et a changé progressivement et de façon significative nos habitudes en termes de production et de conception. Presque toutes les entreprises aujourd&#39;hui, intègrent — ou cherchent à intégrer — d’une manière ou d’une autre, l’IA dans leur processus.</p><p>Mais pendant cette ruée massive vers l’IA, un fait attire mon attention : combien de boites intègrent la technologie du cloud computing dans ses processus comme les milliers d’outils IA utilisés à longueur de journée ?</p><p>Selon une <a href="https://www.agenceecofin.com/telecom/0107-119900-en-afrique-subsaharienne-27-des-dirigeants-d-entreprise-ont-adopte-l-ia-generative-rapport"><strong>étude </strong></a>de l’<strong>agence ecofin</strong> en 2024, en Afrique subsaharienne, une étude indique que 12 % des dirigeants d’entreprise ont adopté le cloud computing dans l’ensemble de leurs opérations, tandis que 38 % l’ont intégré dans une grande partie de leurs activités. Par ailleurs, 32 % des entreprises ont adopté le cloud dans certaines de leurs opérations, et 19 % n’utilisent pas encore ce type de solutions.</p><p>Vu les statistiques c’est dire à quel point, les choses vont “mal et mieux” malgré le nombre croissant des entreprises adoptant le cloud computing, Aha !</p><p>Cet article explore comment exploiter le cloud pour libérer le potentiel de croissance des entreprises, en mettant l’accent sur des concepts clés comme la <strong>virtualisation</strong>, les <strong>services AWS</strong> et les <strong>réseaux privés virtuels</strong> (VPN)</p><blockquote>Treve de barvadage… Allons dans le vif du sujet.</blockquote><p>En Afrique, le cloud computing est devenu un catalyseur de transformation numérique, offrant aux PMEs et startups, des outils autrefois réservés aux grandes entreprises. Avec des fournisseurs comme <a href="http://aws.amazon.com/"><strong>AWS</strong></a>, cette technologie permet de surmonter les défis d’infrastructure, des couts et de sécurité, tout en accélérant l’innovation et la création de valeur.</p><p><strong>A- Le cloud computing pour les <em>nuls</em></strong></p><p>Quand on parle du cloud, pour bon nombre de personnes, c’est “stockage en nuage”. <em>Instant confession </em>: je pensais la même chose il y a une dizaine d’année. Haha !</p><p>La bonne nouvelle c’est que vous n’aviez pas vraiment tord si vous pensiez ainsi. Mais la vérité c’est que le “stockage en nuage’ comme vous l’appelez, fait lui aussi parti des services qui composent la technologie <strong>Cloud</strong>. <strong><em>Spoiler</em></strong> : On en a des milliers de services dans le cloud. Même l’IA s’héberge dans le cloud et utilise des données centralisées dans le cloud. Waouh !</p><p>Il y a pres de deux à trois decennies en arriere, pour que les entreprises creent leur propres logiciels et stocker en interne leurs données utilisateurs, il leur faut investir, acheter et installer des serveurs, configurer des réseaux, et bien d’autres. Ainsi au fur et à mesure que l’entreprise grandit, il lui faut fournir des ressources en matériels informatique plus couteux. Au final, la scalabilité pose problème.</p><p>C’est la que le cloud computing entre en jeu comme solution ultime aux dépenses onéreuses des départements opérationnels (ops) des entreprises.</p><h4>Le cloud permet donc l’accès à distance, via internet aux ressources (stockage, virtualisation, calcul, réseau, IA, IoT, etc) qu’une infrastructure physique local proposerait.</h4><p>C’es t donc un peu le cas de la majeur partie des entreprises aujourd&#39;hui qui utilisent des ressources serveurs à distance pour leur produit et services sans avoir de serveurs physique en interne dans leur locaux.</p><p>C’est tout ? Oui. Le cloud c’est aussi simple que ca. Haha ! Trop simple, pas vrai ?!</p><p><strong>B- Infrastructure sur site VS Infrastructure Cloud</strong></p><p><strong>L’infrastructure sur site</strong> fait référence aux serveurs, aux matériels et aux logiciels qui sont physiquement installés dans les locaux d’une entreprise. Dans ce cas de figure, l’entreprise a un contrôle total sur l’infrastructure, mais son entretien nécessite des couts élevés.</p><p><strong>L’infrastructure cloud</strong> quant à elle, utilise des ressources informatiques fournies par des fournisseurs de services cloud via internet comme (<a href="http://aws.amazon.com/"><strong>AWS</strong></a><strong>, </strong><a href="https://azure.microsoft.com/"><strong>Microsoft Azure</strong></a><strong>, </strong><a href="https://cloud.google.com/"><strong>Google Cloud</strong></a>). Ici, le contrôle de l’infrastructure est partagé avec le fournisseur de services cloud qui gère l’infra sous-jacente.</p><p>Ici, les couts sont généralement basés sur un modelé de paiement à l’utilisation, ce qui permet de réduire les dépenses inutiles et de ne payer que ce que vous utilisez.</p><p><strong>C- Cas Pratiques : comment les entreprises peuvent exploiter le cloud d’Amazon Web Services (</strong><a href="http://aws.amazon.com/"><strong>AWS</strong></a><strong>) pour résoudre des problèmes complexes ?</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*j48lSQU_uqCN6nqWTNeUEw.jpeg" /><figcaption>Image illustrant le cloud d’Amazon Web Service (AWS). Générée par DALL-E</figcaption></figure><p><strong>Problème 1-) : Partage Sécurisé de la Connexion Internet via un VPN (Virtual Private Network) Cloud pour une Entreprise de Conseil</strong></p><p>Dans une société de conseil basée au Maroc, avec 200 employés travaillant à distance dans toute l’Afrique, les employés accédaient à Internet via des réseaux domestiques non sécurisés, exposant les données clients (études de marché sensibles) à des risques de fuites ou d’interceptions.</p><h4>Solution Cloud AWS : Architecture VPN avec Internet Sécurisé</h4><p>L’objectif ici, est de tunneliser tout le trafic Internet<strong> </strong>des employés via le <strong>cloud AWS</strong> pour :</p><ul><li>Chiffrer les données.</li><li>Appliquer des politiques de sécurité (filtrage DNS, blocage de sites).</li><li>Centraliser les logs d’accès.</li></ul><p><strong>Services Amazon Web Services (AWS) Cloud Utilisés</strong> :</p><ol><li><a href="http://aws.amazon.com/vpn/client-vpn"><strong>AWS Client VPN</strong></a> : Pour connecter les utilisateurs distants au <strong>VPC (Virtual Private Cloud)</strong>.</li><li><a href="https://aws.amazon.com/network-firewall/"><strong>AWS Network Firewall</strong></a> : Pour filtrer le trafic sortant (couche applicative).</li><li><a href="https://aws.amazon.com/route53/resolver-dns-firewall/"><strong>Amazon Route 53 DNS Firewall</strong></a> : Bloque les requêtes <strong>DNS</strong> vers des domaines malveillants et les sites de phishing et de cryptomining</li><li><a href="https://aws.amazon.com/fr/cloudwatch/"><strong>Amazon CloudWatch</strong></a> : Surveillance des <strong>logs</strong> de connexion et du trafic.</li></ol><p><strong>Problème 2-) Sécurisation des Transactions Bancaires avec un VPN Cloud (Fintech, Nigeria)</strong></p><p>Une fintech nigériane proposant des services de paiement mobile devait sécuriser l’accès à ses bases de données sensibles (données clients, historiques de transactions) depuis des agences distantes. Une infrastructure <strong>VPN sur site</strong> (pare-feu physique) était coûteuse et complexe à étendre, elle se tourne vers une solution cloud plus optimale.</p><p><strong>Solution AWS</strong> :</p><ul><li><a href="http://aws.amazon.com/vpn/client-vpn"><strong>AWS Client VPN</strong></a> : Déploiement d’un réseau privé virtuel entièrement managé, connectant les agences à un VPC (Virtual Private Cloud) isolé.</li><li><a href="https://aws.amazon.com/fr/blogs/france/limportance-du-chiffrement-comment-aws-vous-aide-a-proteger-vos-donnees/"><strong>Chiffrement AES-256</strong></a> via IPsec, intégré nativement.</li><li><strong>Gestion des accès</strong> via <a href="https://aws.amazon.com/fr/iam/"><strong>IAM </strong></a>(Interdiction des connexions hors Nigeria pour prévenir les attaques géolocalisées).</li></ul><p><strong>Résultat</strong> :</p><ul><li>Coûts réduits de 50 % vs une solution d’infrastructure sur site.</li><li>Conformité <strong>PCI DSS</strong> simplifiée grâce aux <strong>certifications</strong> <strong>AWS</strong>.</li></ul><p><strong>Problème 3-) Virtualisation et Scalabilité pour une Plateforme E-commerce (Côte d’Ivoire)</strong></p><p>Une marketplace ivoirienne subissait des pics de trafic x10 pendant les soldes de fin d’année, entraînant des plantages de serveurs locaux (temps de réponse / latence &gt; 10s).</p><p><strong>Solution AWS</strong> :</p><ul><li><strong>Auto Scaling Group (</strong><a href="https://aws.amazon.com/fr/ec2/"><strong>Amazon EC2</strong></a><strong>)</strong> : Ajout automatique d’instances t4g.medium pendant les pics, basé sur des métriques <a href="https://aws.amazon.com/fr/cloudwatch/"><strong>CloudWatch </strong></a>(CPU &gt; 70 %).</li><li><strong>Conteneurisation (</strong><a href="https://docs.aws.amazon.com/AmazonECS/latest/developerguide/AWS_Fargate.html"><strong>ECS Fargate</strong></a><strong>)</strong> : Migration des microservices vers des conteneurs serverless, éliminant la gestion des serveurs.</li><li><a href="https://aws.amazon.com/rds/aurora/"><strong>Amazon Aurora</strong></a> : Base de données MySQL-compatible avec réplication Multi-AZ, garantissant une latence &lt; 3 ms.</li></ul><p><strong>Résultat</strong> :</p><ul><li>Gestion de 50 000 requêtes/minute pendant les soldes, sans interruption.</li><li>Coûts optimisés grâce à l’arrêt automatique des instances inutilisées.</li></ul><p><strong>Probleme 4-) Analyse Prédictive pour l’Agriculture (Kenya) — IA et Stockage</strong></p><p>Une coopérative agricole voulait prédire les rendements des cultures en analysant des images satellites (10 To/mois) et des données météo. Les serveurs locaux manquaient de puissance de calcul.</p><p><strong>Solution AWS</strong> :</p><ul><li><strong>Stockage</strong> : Archivage des images dans <a href="https://aws.amazon.com/s3/storage-classes/glacier/"><strong>Amazon S3 Glacier</strong></a> (0,004 $/Go/mois).</li><li><strong>Calcul</strong> : Utilisation de <a href="https://aws.amazon.com/fr/ec2/"><strong>EC2 </strong></a><strong>P3</strong> (GPU NVIDIA V100) pour entraîner un modèle de vision par ordinateur via <a href="https://aws.amazon.com/sagemaker/"><strong>SageMaker</strong></a>.</li><li><strong>Pipeline de données</strong> : <a href="https://aws.amazon.com/glue/"><strong>AWS Glue</strong></a> pour nettoyer et préparer les données météo (format CSV/JSON).</li></ul><p><strong>Résultat</strong> :</p><ul><li>Précision de 92 % sur les prédictions de rendement, permettant une optimisation des récoltes.</li><li>Réduction des coûts de stockage de 70 % vs une solution d’infrastructure sur site.</li></ul><p><strong>Problème 5-) Télémédecine et Streaming Vidéo (Afrique du Sud) — Réseau et Calcul dans le Cloud</strong></p><p>Une startup de télémédecine avait des difficultés à diffuser des consultations vidéo HD en temps réel vers des zones rurales, avec une latence élevée (&gt;500 ms).</p><p><strong>Solution AWS</strong> :</p><ul><li><a href="https://aws.amazon.com/fr/ivs/"><strong>Amazon IVS </strong></a><strong>(Interactive Video Service)</strong> : Service managé pour le streaming low-latency (&lt; 3s), intégrant CDN via <a href="https://aws.amazon.com/fr/cloudfront/"><strong>CloudFront</strong></a>.</li><li><a href="https://aws.amazon.com/global-accelerator/"><strong>AWS Global Accelerator</strong></a> : Routage optimisé du trafic vers le point de présence AWS le plus proche (ex. Johannesburg).</li><li><a href="https://aws.amazon.com/lambda/"><strong>AWS Lambda</strong></a> : Déclenchement de workflows automatisés (ex. envoi de rapports médicaux après chaque consultation).</li></ul><p><strong>Résultat</strong> :</p><ul><li>Diffusion vidéo en 1080p avec une latence &lt;200 ms, même en 3G.</li><li>30 000 consultations/mois sans investissement en matériel vidéo.</li></ul><blockquote>Et une dernière pour la fin…</blockquote><p><strong>Probleme 6-) Gestion de Flottes Logistiques (Sénégal) — IoT et Virtualisation</strong></p><p>Une entreprise de logistique devait suivre 500 camions en temps réel, mais ses serveurs locaux ne peuvent gérer le volume de données IoT (GPS, capteurs de température).</p><p><strong>Solution AWS</strong> :</p><ul><li><a href="https://aws.amazon.com/iot-core/"><strong>AWS IoT Core</strong></a> : Collecte et traitement de 10 000 messages/minute depuis les capteurs via <a href="https://mqtt.org/"><strong>MQTT</strong></a>.</li><li><strong>Virtualisation des véhicules</strong> : Création de “<a href="https://aws.amazon.com/fr/what-is/digital-twin/#:~:text=Les%20jumeaux%20num%C3%A9riques%20s&#39;appuient,%C3%A0%20jour%20en%20temps%20r%C3%A9el."><strong>jumeaux numériques</strong></a>” dans <a href="https://aws.amazon.com/dynamodb/"><strong>DynamoDB</strong></a>, reflétant l’état réel des camions (ex. température, localisation).</li><li><a href="https://aws.amazon.com/kinesis/"><strong>Amazon Kinesis Data Analytics</strong></a> : Détection d’anomalies (ex. déviation d’itinéraire) en temps réel.</li></ul><p><strong>Résultat</strong> :</p><ul><li>Réduction de 20 % de la consommation de carburant grâce à l’optimisation des routes.</li><li>Alertes temps réel en cas de rupture de la chaîne du froid.</li></ul><p>De l’IA, au streaming, en passant par l’IoT, le calcul, le stockage et les réseaux, nous avions vu par ces cas de résolutions de problèmes, comment la technologie du cloud computing à travers les fournisseurs cloud et leurs services, peut être un outil stratégique pour les entreprises et les startups, tout comme l’IA Générative l’est aujourd&#39;hui.</p><p>Durant cet article, je me suis expressément focalisé sur les services d”Amazon (<strong>AWS Cloud</strong>). N’oublions néanmoins donc pas les autres fournisseurs dont Google Cloud, Microsoft Azure et bien d’autres.</p><h4><a href="#"><strong>Epilogue:</strong></a><strong> Vous êtes une entreprise, une startup ou un particulier ayant activement besoin d’un Ingénieur Cloud DevOps, pour utiliser et intégrer efficacement le cloud dans vos process, discutons-en ici via mail : <em>florentallagbe229@gmail.com ou </em></strong><a href="https://linkedin.com/in/orpheric-allagbe-1863961b6">sur mon site</a>.</h4><h4>J’enseigne et je forme également vos équipes IT, aux bonnes pratiques DevOps.</h4><h4><strong>Je suis Orphéric Florent,</strong></h4><h4><strong>AWS Cloud Technology Consultant Certified.<br>AWS Cloud Support Associate Certified.</strong></h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=23cd51799cd3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to create a backup of your Laravel application on AWS S3?]]></title>
            <link>https://medium.com/@Orpheric_Codeur/how-to-create-a-backup-of-your-laravel-application-on-aws-s3-5b438dcc7424?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/5b438dcc7424</guid>
            <category><![CDATA[laravel-application]]></category>
            <category><![CDATA[aws-s3]]></category>
            <category><![CDATA[backup]]></category>
            <category><![CDATA[backup-and-restore]]></category>
            <category><![CDATA[aws-cloud]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Mon, 22 Apr 2024 15:48:41 GMT</pubDate>
            <atom:updated>2024-04-22T16:18:00.757Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/627/1*nGX9KzDjkZTSfTa3dAzGVQ.png" /></figure><p><strong>Context:</strong> You have a <strong>SaaS </strong>solution developed in the form of a <strong>web application </strong>designed with <strong>Laravel</strong> technology.</p><p>Having a data backup of your application is therefore crucial because it provides an extra layer of protection against accidental loss or “corruption” of your application data.<br>And if you thought that data leaks only happen to others, remember that even the most robust and secure applications are also victims of unforeseen events such as hardware failure, infrastructure failures or even human errors and attacks.</p><p>So in such cases, having a backup can mean the difference between rapid recovery and catastrophic loss.</p><p>Also, having a backup is very important because it is essential for your recovery after a disaster or unfortunate event on your cloud infrastructure. And in the unfortunate event of a security breach, a backup can help you restore your application data to a lower state, thereby reducing the risk of data loss and minimizing the impact on your SaaS solution and therefore your business. your business.</p><p><strong>Backups</strong> therefore intervene not only as a preventive measure but also as a curative measure when the “damage” may already be done.</p><p>Before moving on to the technical part, let’s quickly review <strong>five (05)</strong> <strong>principles to respect</strong> to set up the backups of your web applications.</p><p><strong>1-</strong> Always store your backups on external storage such as S3. It is STRONGLY NOT RECOMMENDED TO STORE YOUR BACKUPS ON THE SAME SERVER AS YOUR APPLICATION. It’s not a good idea.</p><p><strong>2-</strong> Have regular backups. That’s right, it all depends on the nature of your application and the money you spend on cloud storage. However, having at least one backup per day is the minimum you should do.</p><p><strong>3-</strong> Keep at least one week of backups in your storage. Note that this always depends on your application. But in case you encounter a bug on your infrastructure that hampers your data, you will have at least a week of backup to recover. This principle is therefore only the result of the previous one.</p><p><strong>4-</strong> Because backup means <strong>restoration</strong>, have a restoration script that can be executed at any time. Ideally, the entire process should be automated and in this case all you need to do is run the script or press a button.</p><p><strong>5-</strong> Always include <strong>database dump</strong>, <strong>redis dump</strong>, <strong>storage folder</strong> and <strong>.env file</strong> in the <strong>backup</strong>. However, does not include the “<strong>vendor</strong>” or “<strong>node_modules</strong>” folders in it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*yqqoZc_-4BheB5aBfZN7RQ.jpeg" /></figure><p>The following technical part will revolve around <a href="https://aws.amazon.com/s3/?nc1=h_ls"><strong>AWS S3</strong></a>, one of the best ways to store backups. <strong>AWS S3 (Simple Storage Service)</strong> is a highly scalable, secure and durable object storage service provided by Amazon Web Services (AWS) which allows us to store and retrieve any amount of data from anywhere on the web using a simple <strong>HTTP API</strong>.</p><p>To store and organize this data, <strong>AWS S3</strong> is composed of Buckets which are not folders but rather resemble a kind of <strong>git repository </strong>where we can deposit numerous folders and files. Literally, AWS S3 <strong>buckets </strong>are object containers.<br><strong>Objects </strong>are therefore individual <strong>files stored</strong> in <strong>buckets</strong>. These <strong>objects</strong> can be from a simple text file to a large video file or even a database backup. Note that each Object is identified by a unique key which is used to retrieve the object from the <strong>bucket</strong>.</p><p>GOOD ! No more chatter… Ha ha. We now know more about <strong>backups </strong>and the principles that govern them, as well as A<strong>WS S3</strong>. NOW; Let’s Goooooo!</p><p><strong>PHASE A: </strong>AWS Configuration</p><p>Before starting the configurations, go to your AWS space, search for “<strong>S3</strong>” in the search bar then in the left side panel, click on <strong>Buckets </strong>to create one. We will name it “<strong>laravel-saas-backups</strong>”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HyfRJRmnblDHcjXjp9MXRg.png" /></figure><p>Once your bucket is created, try loading a simple zip file as an object in your Bucket, as a test</p><p>S<strong>tep 1:</strong> Install the AWS CLI tool. Depending on your operating system, consult this documentation <a href="http://docs.aws.amazon.com/cli/latest/userguide/getting-started-install.html"><strong>here</strong></a><strong> </strong>for installation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/489/1*IGrSCSTM7C8UWn32uXWjjg.png" /></figure><p><strong>Step2: </strong>Once the installation is complete, go to your machine’s command line interface and type: <strong>aws — version</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/525/1*v-2j7ERZ3UO4bOOq5HzbHQ.png" /></figure><p><strong>Step 3: </strong>Search for the access key for authorization. Here, you need to connect to your AWS Cloud account space or create an account if you have not already done so.<br>Once in your space, click on your username in the upper right corner and click on “<strong>security credentials</strong>”. You will then find an “<strong>access key</strong>” section to create one. Resources available <a href="http://docs.aws.amazon.com/singlesignon/latest/userguide/get-set-up-for-idc.html"><strong>here</strong></a></p><p><strong>Note:</strong> However, if you are just testing AWS, you can use a root user access key. However, this is strongly discouraged if you are on your company’s server infrastructure or in a production environment.</p><p><strong>Step 4: </strong>We will configure our CLI by running this command: aws configure. Following this command, you will have to enter your access key, your secret key, your default region, etc.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/337/1*1G0DuYTHvIRDqgniL4lP-A.png" /></figure><p><strong>Step5:</strong> Once these configurations are completed, we can enter the following command to display the klist of all the buckets present on our AWS S3 space: <strong>aws s3 ls</strong></p><p>The output should show the bucket name: “<strong>laravel-sass-backups</strong>” previously created at the top. We can also download the specific objects (files) placed on our buckets using the command: &lt;&lt; <strong>aws s3 cp</strong> “<strong>URL-S3 of the object</strong>” “<strong>./path”</strong>&gt;&gt; . This <strong>command</strong> will be used in particular during the <strong>restoration</strong> <strong>process</strong></p><p><strong>PHASE B:</strong> Configuring our Laravel Application for backup</p><p>Since this article is mainly about setting up backups of a Laravel application on AWS S3, we move on to the following different steps:</p><p><strong>Step1:</strong> Set the following environment variables in order to use S3 in our Laravel application</p><pre>AWS_ACCESS_KEY_ID=<br>AWS_SECRET_ACCESS_KEY=<br>AWS_DEFAULT_REGION=eu-north-1<br>AWS_BUCKET=laravel-sass-backups<br>AWS_USE_PATH_STYLE_ENDPOINT=false</pre><p><strong>NOTE: </strong>Please enter your access key, secret key and corresponding region</p><p><strong>Step2: </strong>Install this package</p><pre>composer require league/flysystem-aws-s3-v3 &quot;^3.0&quot;</pre><p>The <strong>league/flysystem-aws-s3-v3</strong> package makes it easy to integrate and manipulate files stored on Amazon S3 into your PHP/LARAVEL applications using the <strong>Flysystem file system</strong> and the <strong>AWS PHP v3 SDK</strong>.</p><p><strong>Step3: </strong>Installation of the “<strong>spatie/laravel-backup</strong>” package which allows you to create backups from a Laravel application. Check out the documentation <a href="http://spatie.be/docs/laravel-backup/v8/installation-and-setup"><strong>here</strong></a></p><p>After installing “<strong>spatie/laravel-backup</strong>” and publishing the configuration file “<strong>config/backup.php</strong>” by strictly following the documentation steps, go to the next step</p><p><strong>Step4: </strong>Configure the destination path in “<strong>config/backup.php</strong>”</p><pre>             /*<br>             * The disk names on which the backups will be stored.<br>             */<br>            &#39;disks&#39; =&gt; [<br>                &#39;s3&#39;,<br><br>            ],</pre><p><strong>Step5:</strong> Now let’s automate the backup of our application using these two commands in “A<strong>pp/Console/Kernel.php</strong>”</p><pre>    protected function schedule(Schedule $schedule): void<br>    {<br>        $schedule-&gt;command(&#39;backup:clean&#39;)-&gt;daily()-&gt;at(&#39;01:00&#39;);<br>        <br>        $schedule-&gt;command(&#39;backup:run&#39;)-&gt;daily()-&gt;at(&#39;01:30&#39;);<br>    }</pre><p>These lines of code define two scheduled tasks in Laravel. The first task (<strong>backup:clean</strong>), which will delete old backups, is scheduled to run every day at 1:00 a.m., while the second task (<strong>backup:run</strong>), which will create a new backup, is scheduled to run every day. at 1:30 a.m.<br>It will thus create a zip file including our databases and the rest of our application directory.</p><p>While waiting to see the automated backup, to have an overview of our backup, let’s manually execute <strong>backup:run</strong> by following the following command.</p><pre>php artisan backup:run</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/776/1*zh1nCia2UcJM2rw6Hud-yg.png" /></figure><p>And now check your AWS S3 bucket, your new backup is there perfectly. Haha…</p><p>Note that the manual backup was done just to illustrate what will happen when the scheduled backup tasks run at the mentioned time. So no need to do it manually in your production application.</p><p><strong>Note: </strong>You would have undoubtedly noticed that on your A<strong>WS S3 bucket</strong>, your backup in zip format is in a directory bearing the name of your local Laravel application, the one entered in the “APP_NAME” variable of the .env file. To change this if you want, you can reconfigure everything in the “<strong>config/backup.php</strong>” file. Don’t forget the documentation above on “s<strong>patie/laravel-backup</strong>” to guide you.</p><p>Since there is no backup without restoration, I will discuss this in a future article, if you are interested.</p><p><strong>EPILOGUE: </strong>To set up the <strong>backup</strong> of your <strong>web applications</strong>, you are not obliged to use <strong>AWS S3</strong>, if you do not want to because of its somewhat ugly UI or for other personal reasons. The <strong>cloud provider</strong> <strong>DigitalOcean </strong>through <a href="https://www.digitalocean.com/products/spaces"><strong>Digital Ocean Spaces</strong></a> offers an <strong>S3 compatible </strong>storage solution. S3 compatible means it has the same <strong>API</strong> as <strong>AWS</strong>.<br>If you’re interested, let me know and I’ll probably talk about it in another article. THANKS !</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KsuXQm1gEJZrEVcy8UkNng.png" /></figure><p>Here we are at the end of this all-in-one article. Now, you had the means to make regular backups on your Laravel web applications.<br>Don’t hesitate to leave me your feedback.</p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><em>I am </em><strong><em>Orphéric</em></strong><em>.<br></em><strong><em>Laravel Software Developer</em></strong><em> and </em><strong><em>Data Infrastructure Engineer</em></strong><em>.</em></p><p>Leave me a message <a href="/service/mailto:florentallagbe229@gmail.com"><strong>here</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5b438dcc7424" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to optimize your Nginx server in terms of the number of requests/trafics processed…]]></title>
            <link>https://medium.com/@Orpheric_Codeur/how-to-optimize-your-nginx-server-in-terms-of-the-number-of-requests-trafics-processed-b3bbb0edb5a7?source=rss-ba8df5dc9a11------2</link>
            <guid isPermaLink="false">https://medium.com/p/b3bbb0edb5a7</guid>
            <category><![CDATA[digital-ocean-droplet]]></category>
            <category><![CDATA[nginx-worker-connections]]></category>
            <category><![CDATA[nginx]]></category>
            <category><![CDATA[nginx-optimization]]></category>
            <category><![CDATA[nginx-worker-process]]></category>
            <dc:creator><![CDATA[Orphéric ALLAGBE]]></dc:creator>
            <pubDate>Wed, 10 Apr 2024 23:47:38 GMT</pubDate>
            <atom:updated>2024-04-10T23:47:38.690Z</atom:updated>
            <content:encoded><![CDATA[<h3>How to optimize your Nginx server in terms of the number of requests/trafics processed simultaneously in your web app ?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YhxPmAYI5ZNptKSS5qUnMw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/410/1*TM3AqVx_Jdymc7qGaYDb5Q.jpeg" /></figure><p><strong>Context: </strong>You have your web application or your SaaS solution running on the cloud using the popular NGINX server. And according to your data you have a server with four (04) cores in its CPU with an average of 4000 simultaneous user connections.<br>Your application being scalable and responding to a highly demanded need, you go from 4000 active users to almost 10,000 or even 100,000 users. Consequence: more gains but also and above all INCREASED PERFORMANCE PROBLEMS or SLOW CONNECTIONS to your server and consequently to your web application.</p><p>In the remainder of this article, we will quickly try to resolve this performance problem on our <strong>NGINX</strong> <strong>server</strong>. Here we go !</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/380/1*Vu_p8NYOF3YhDIYVvhbXfg.png" /></figure><p>Consider the snippet of code from the NGINX configuration file: nginx.conf located at /etc/nginx/nginx.conf.</p><pre>user www-data;<br>worker_processes auto;<br><br>events {<br>  worker_connections 1024;<br>}</pre><p>Here, “worker_processes” controls the number of worker processes that can use NGINX. Its value is generally equal to the number of CPU processors available on our server. The “auto” attribute does this automatically in our case.</p><p>“worker_connections” defines the number of connections that a worker process can open. To find its value, type in the command: ulimit -n. The latter gives us the number of file descriptors that a worker process can open under Linux. It is often 1024.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1018/1*vmf2Y7bnZf9BYmFNo7IuAg.png" /></figure><p>To summarize, considering the problem posed in our context at the top of this article, if your server has (04) cores in your CPU and you define “work_connections” to 1024 based on the value returned by “ulimit -n” of course, this means that your NGINX server can handle 4096 (1024 * 4) connections (users) at the same time, simultaneously.</p><p>So, if the 4097th user arrives, his connection will be placed in a queue, and will be served only when NGINX has some space to breathe. Otherwise, the connection of 4097 ieme will be almost slow.</p><p>Let’s assume that you have just realized through the figures that your application is now holding traffic of almost 100,000 connections, which has led to performance problems and incessant complaints from users. We will try to resolve this.</p><p><strong>Warning:</strong> if you do not have high traffic peaks or performance problems, we do not recommend that you make all of the settings that follow. This is for the good of your cloud server infrastructure.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/275/1*ppaNs_4TB6XnRluhc8EqGQ.jpeg" /></figure><p>You probably remember “<strong>ulimit -n</strong>”, this command which allowed us to know the value assigned to “worker_connections”. In reality, “ulimit -n” sets the soft, default limit of connections that a worker process can open.<br>And if there is a soft limit, there must be a stricter limit somewhere, right? This brings us back to “<strong>ulimit -Hn</strong>” which defines the strict and REAL limit of connections that a process can open.</p><p>The “<strong>ulimit -Hn</strong>” command gives us <strong>1,048,576</strong> in our case. The soft limit, by default being only <strong>1024</strong>, we can therefore easily upgrade or bypass to have a number allowing us to better manage the excess of incoming traffic.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1011/1*tG3f09VI2KfBdyjMFUt7YQ.png" /></figure><p>Considering the strict limit and in order to fill the surplus of simultaneous traffic going from 4000 to 100,000 or so, we had the following configuration</p><pre>user www-data;<br>worker_processus auto;<br>worker_rlimit_nofile 25000;<br><br>events {<br>  worker_connections 25000;<br>}</pre><p>You would have noticed that we had a slight upheaval in the new code. “worker_rlimit_nofile” defines the maximum number of file descriptors in our system. Its value therefore corresponds exactly to that of “ulimit -n” corresponding to the default soft limit of connections that a SINGLE process can open.</p><p>For a quick calculation, if in our case we wanted our server to be able to process 100,000 user connections simultaneously, we will need a process to be able to process 25,000 connections alone due to (04) process cores on our server.<br>Which gives us: “<strong>worker_rlimit_nofile === ulimit === 25000</strong>”</p><p>Our strict limit being <strong>1,048,576</strong> per process we can therefore validly afford to be satisfied with 25,000 connections per process (04) or the 100,000 sought.</p><p>NB1: IT IS IMPORTANT TO NOTE THAT A SETTING OF “<strong>work_rlimit_nofile</strong>” TOO HIGH can cause performance problems even if the server does not have enough resources to handle the load.</p><p>NB2: ONLY MODIFY THIS WHEN YOU EXPERIENCE REAL TRAFFIC PEAKS OR PROBLEMS.</p><p>THANKS ! I hope you enjoyed this article.<br>Don’t hesitate to leave me a comment if you have any concerns.</p><p><em>Signed Orphéric.<br>I’m Laravel Software Developer and Data Infrastructure Engineer</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b3bbb0edb5a7" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>