<?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 Aditya Soni on Medium]]></title>
        <description><![CDATA[Stories by Aditya Soni on Medium]]></description>
        <link>https://medium.com/@adityasonittyl?source=rss-68478c793efe------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*F_PlT7MDZwHfLv6TLlJqiw.jpeg</url>
            <title>Stories by Aditya Soni on Medium</title>
            <link>https://medium.com/@adityasonittyl?source=rss-68478c793efe------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 08:57:48 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@adityasonittyl/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[What is FinOps? Why Cloud/DevOps/SRE Engineers Should Care About It!]]></title>
            <link>https://adityasonittyl.medium.com/what-is-the-finops-why-cloud-devops-sre-engineers-should-care-about-it-225d2c987a36?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/225d2c987a36</guid>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[finops]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[cloud]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Fri, 16 Jun 2023 14:34:49 GMT</pubDate>
            <atom:updated>2023-06-17T03:55:44.424Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oaQxfmbHn416W1Q1Ee2glA.jpeg" /></figure><p>You must have heard of <em>Agile and DevOps </em>methodologies, but wait have you been introduced to <strong>FinOps Lifecycle?<br></strong><em>If NOT read till the end!</em></p><h4><strong>What is FinOps:</strong></h4><p>FinOps is an evolving cloud financial management discipline and cultural practice that enables organizations to get maximum business value by helping <em>engineering, finance &amp; business teams to collaborate on a data-driven spending decision.</em></p><h4><strong>What is FinOps Lifecycle/Framework:</strong></h4><p>FinOps, which stands for Financial Operations. <strong>It’s just like Agile and DevOps methodologies</strong>, the FinOps lifecycle follows a cyclical approach to cloud cost management and optimization.</p><p>It involves continuous monitoring, analysis, and optimization of cloud costs throughout the lifecycle of cloud adoption and emphasizes collaboration between finance, technology, and business teams with, iterative improvement of cost management practices.</p><p>Or let&#39;s say — C<em>ollaboration between business and engineering teams.</em></p><blockquote><em>FinOps isn’t about saving money rather than. FinOps is about making money!</em></blockquote><h4><strong>Understanding the FinOps Lifecycle:</strong></h4><p>The FinOps Lifecycle follows a continuous and iterative process to effectively manage and optimize cloud costs. It consists of three core phases: <strong>Inform, Optimize, and Operate</strong>. <br>Let’s delve into each phase:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/776/1*lhxd3XhSg4N_Ic5mlzga6g.png" /><figcaption>Image credit: FinOps Foundation</figcaption></figure><h4><strong>Inform Phase:</strong></h4><p>The Inform phase focuses on gathering data, analyzing costs, and providing insights to stakeholders. It involves establishing cost visibility, identifying cost drivers, and promoting a culture of cost awareness.</p><p>Examples of activities in the Inform phase include:</p><ul><li>Cost Reporting</li><li>Cost Allocation</li><li>Showback/Chargeback</li><li>Cost Optimization Recommendations</li></ul><h4><strong>Optimize Phase:</strong></h4><p>The Optimize phase focuses on implementing cost-saving measures and driving efficiency in cloud resource utilization. It involves identifying areas of improvement, implementing optimization strategies, and tracking the impact of those actions.</p><p>Examples of activities in the Optimize phase include:</p><ul><li>Rightsizing</li><li>Spot Instances</li><li>Automation and Orchestration</li><li>Reserved Instances/Savings Plans</li></ul><h4><strong>Operate Phase:</strong></h4><p>The Operate phase focuses on ongoing cost management and governance practices. It involves establishing policies, monitoring costs, and ensuring compliance with cost targets and budgetary constraints.</p><p>Examples of activities in the Operate phase include:</p><ul><li>Cost Governance</li><li>Cost Anomaly Detection</li><li>Cost Optimization Reviews</li><li>Cost Forecasting</li></ul><p>It’s important to note that these phases are not strictly linear, but rather iterative. Organizations continuously cycle through these phases, refining their cost management practices and optimizing their cloud spending over time.</p><h4><strong>FinOps Principles:</strong></h4><p>Here are the FinOps Principles!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u6YEh8mNkXEMPRBYNxZSJA.png" /><figcaption>Image credit: FinOps Foundation</figcaption></figure><h4><strong>FinOps Maturity Model:</strong></h4><p>FinOps Maturity Model is a framework that helps organizations assess their level of maturity in implementing FinOps practices!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/398/1*Z-_GnXOP7wQNd1C-u0VzNQ.png" /><figcaption>Image credit: FinOps Foundation</figcaption></figure><p><strong>[A] Crawl Phase:</strong> In the Crawl phase, organizations are in the early stages of adopting FinOps practices. They are focused on establishing cost visibility, gathering data, and building basic cost reporting capabilities. The primary goal is to gain an understanding of cloud costs and begin the journey toward cost optimization.</p><p><strong>[B] Walk Phase:</strong> In the Walk phase, organizations have made progress in their FinOps journey. They have implemented more advanced cost management practices, such as cost allocation, show back/chargeback, and optimization recommendations. They are actively monitoring and analyzing costs to identify opportunities for optimization.</p><p><strong>[C] Run Phase:</strong> In the Run phase, organizations have achieved a higher level of maturity in their FinOps practices. They have implemented automation, advanced optimization strategies, and robust cost governance policies. They have established cross-functional collaboration between finance, technology, and business teams to drive cost optimization initiatives.</p><blockquote>FinOps Maturity Model is not a one-size-fits-all approach. Each organization’s journey may differ based on their specific goals, resources, and cloud adoption stage.</blockquote><h4><strong>FinOps Team Structures:</strong></h4><p>The below diagram demonstrates how organizations operate on the FinOps model!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*veTTqlzaro2AiruKpqImJw.png" /><figcaption>Image credit: FinOps Foundation</figcaption></figure><h4>Why Cloud/DevOps/SRE Engineers Should Care About It:</h4><ol><li>It helps optimize cloud costs, ensuring efficient resource utilization and eliminating wasteful spending.</li><li>It enables better alignment between technology and business goals, leading to more effective decision-making and resource allocation.</li><li>It fosters a culture of cost accountability and transparency, empowering engineers to make informed decisions that drive financial efficiency.</li></ol><h4><strong>Conclusion:</strong></h4><p>Embrace FinOps, drive cost efficiency, and propel your DevOps and cloud initiatives to new heights!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*s0dyfxCWj_Ru72G88xF-rQ.png" /><figcaption>Image credit: FinOps Foundation</figcaption></figure><p>That’s all for this time. <br>In case you’re looking to adapt FinOps in your organization check out the FinOps Foundation <a href="https://www.finops.org/wg/adopting-finops/">official documentation</a>!</p><p><a href="https://medium.com/u/68478c793efe">Aditya Soni</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=225d2c987a36" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Humans are not replaceable in the Artificial Intelligence (AI) revolution !!]]></title>
            <link>https://adityasonittyl.medium.com/humans-are-not-replaceable-in-the-artificial-intelligence-ai-revolution-48655b46c38f?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/48655b46c38f</guid>
            <category><![CDATA[future]]></category>
            <category><![CDATA[humanity]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Tue, 06 Jun 2023 07:12:40 GMT</pubDate>
            <atom:updated>2023-06-06T07:20:32.877Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Humans are not replaceable in the </strong>Artificial Intelligence (AI)<strong> revolution!!</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OehAO9DLcEXLSqrKJRJ3xQ.png" /></figure><p>You might think that sooner or later AI will replace you, your jobs, or maybe humankind itself if you’ve ever thought about the possibility, read this blog until the end!</p><blockquote>Artificial Intelligence (AI) tools like <strong>ChatGPT</strong> are game-changing for any industry, but when compared with human beings, <strong>they have a lot of limitations</strong>!!</blockquote><p>While AI has indeed achieved incredible feats, it is essential to recognize that creativity and human-like thinking are complex aspects that AI struggles to replicate.</p><h4>Let’s understand the same</h4><blockquote><strong>Don’t forget our creativity and thinking power have enabled us to make and develop a number of AI tools to make our lives easier, IT’S US WHO MADE THEM !!</strong></blockquote><p>There are several limitations that AI currently faces compared to humans.</p><ol><li><strong>Lack of Creativity &amp; Thinking:</strong> AI algorithms are trained on vast amounts of data, but <em>struggle to generate truly original and imaginative ideas</em> and human beings.</li><li><strong>Emotional Intelligence: </strong>AI is for humans and but they <em>lack the empathy and emotional intelligence</em> that humans naturally possess.</li><li><strong>Real-World Experience:</strong> AI knows what you feed them and learn from large datasets, they <em>do not possess personal experiences, intuition</em>, or subjective perspectives that humans bring to the table.</li><li><strong>Lack of Common Sense:</strong> <em>If you feed AI 2+2=6 it believes and knows that only</em> we humans have a wealth of common sense knowledge that allows them to navigate the world and make logical deductions effortlessly.</li><li><strong>Ethical Decision-Making:</strong> AI systems lack moral and ethical reasoning abilities. <em>They rely solely on the rules and objectives defined by their creators.</em></li><li><strong>Adaptability</strong>: Humans possess an inherent ability to adapt to new situations, learn from them, and generalize that knowledge to different scenarios. <em>AI systems typically require extensive retraining to adapt to new tasks or environments</em>, making them less flexible compared to humans.</li><li><strong>Physical Limitations:</strong> <em>Without human existence who will use AI or it’s worth exiting?</em> AI exists purely in the digital realm and does not possess physical senses.</li></ol><p>Not today but tomorrow these capabilities of AI may enhance but again we have to keep upskilling ourselves and our creativity, then only we can be one step ahead of failing soon one great way to do this is to collaborate with AI.</p><blockquote><em>NOT DEPEND ON AI, BUT USE IT TO TRAIN YOUR MIND !!</em></blockquote><p><em>until next time see you</em></p><p>@adityasonittyl</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=48655b46c38f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What Are the 12-Factor Principles of Cloud-Native Microservices?]]></title>
            <link>https://medium.com/cloud-native-daily/what-are-12-factor-principles-of-cloud-native-microservices-eb25e95e5906?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/eb25e95e5906</guid>
            <category><![CDATA[cloud]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[cloud-native]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[microservices]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Mon, 29 May 2023 14:18:30 GMT</pubDate>
            <atom:updated>2023-06-06T01:19:17.640Z</atom:updated>
            <content:encoded><![CDATA[<h4>Discussing the 12-factor principles and delving into their significance in building successful microservices.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NLSsLmUzorCVdadxBfjCEw.png" /></figure><p>You must have heard the word Cloud-Native &amp; microservices before, but do you really know the best practices while configuring or migrating your applications from monolithic to microservices?</p><blockquote>Today microservices is the term so common for any developer who is building applications, or Cloud/DevOps/SRE engineer who deploys or manages applications.</blockquote><p>Microservices architecture has revolutionized the way we build and deploy modern applications. It offers benefits such as scalability, resilience, and flexibility. However, harnessing the full potential of microservices requires adhering to proven best practices. One such set of principles that has gained widespread recognition is the 12 Factor Principles.</p><p>Let’s discuss the 12 Factor Principles and delve into their significance in building successful microservices.</p><h4><strong>1. Single Codebase</strong></h4><p>Maintaining a single codebase per application ensures consistency and simplicity. It eliminates complexities arising from managing multiple repositories and facilitates efficient collaboration among development teams.</p><blockquote>For example, a developer has some commits not yet deployed to staging; staging has some commits not yet deployed to production. But they all share the same codebase, thus making them identifiable as different deploys of the same app.</blockquote><h4><strong>2. Explicit Dependencies</strong></h4><p>By explicitly declaring and isolating dependencies, microservices become more self-contained and independent. This principle ensures clear boundaries between components, making it easier to manage and update dependencies.</p><blockquote>For example, the build command for Ruby/Bundler is bundle install, while for Clojure/Leiningen it is lein deps.</blockquote><h4><strong>3. Configurations Made Easy</strong></h4><p>Storing configurations separately from the code enables easy customization and deployment across different environments. It allows for swift configuration changes without requiring code modifications, promoting agility and flexibility.</p><h4><strong>4. Backing Services Unleashed</strong></h4><p>Backing services, such as databases or message queues, should be treated as attached resources. This principle emphasizes abstracting service dependencies and employing proper abstractions to leverage different services seamlessly.</p><blockquote>For example, if the app’s database is misbehaving due to a hardware issue, the app’s administrator might spin up a new database server restored from a recent backup. The current production database could be detached, and the new database attached — all without any code changes.</blockquote><h4><strong>5. Smooth Build, Release, and Run</strong></h4><p>Strictly separating the build, release, and run stages streamlines the deployment process. It enables reproducibility, and versioning, and promotes automation, ensuring that what works in development is identical in production.</p><h4><strong>6. Processes as First-Class Citizens</strong></h4><p>Treating processes as first-class citizens allow for better manageability and scalability. Microservices should be designed to be stateless and not rely on long-lived processes. This enables easy scaling, deployment, and recovery of microservices.</p><blockquote>For example, downloading a large file, operating on it, and storing the results of the operation in the database. The twelve-factor app never assumes that anything cached in memory or on disk will be available on a future request or job — with many processes of each type running, chances are high that a future request will be served by a different process.</blockquote><h4><strong>7. Port Binding</strong></h4><p>Microservices should export services via a well-defined port binding mechanism. This allows services to communicate with each other using standardized protocols, enabling seamless integration and interoperability.</p><blockquote>For example, PHP apps might run as a module inside <a href="http://httpd.apache.org/">Apache HTTPD</a>, or Java apps might run inside <a href="http://tomcat.apache.org/">Tomcat</a>.</blockquote><h4><strong>8. Concurrency</strong></h4><p>Microservices should be designed to embrace concurrency, allowing them to handle multiple requests simultaneously. This ensures optimal utilization of resources and improves the overall performance and responsiveness of the system.</p><blockquote>For example, HTTP requests may be handled by a web process, and long-running background tasks handled by a worker process.</blockquote><h4><strong>9. Disposability</strong></h4><p>Designing microservices to be disposable enables easy startup and shutdown, facilitating rapid scaling and deployment. Microservices should be able to be created, terminated, and restarted without causing disruptions to the system.</p><blockquote>For example, on <a href="http://www.rabbitmq.com/">RabbitMQ</a> the worker can send a <a href="http://www.rabbitmq.com/amqp-0-9-1-quickref.html#basic.nack">NACK</a>; on <a href="https://beanstalkd.github.io/">Beanstalkd</a>, the job is returned to the queue automatically whenever a worker disconnects.</blockquote><h4><strong>10. Dev/Prod Parity</strong></h4><p>Ensuring development and production environments are as similar as possible minimizes deployment-related issues. By leveraging containerization and automation, the differences between environments can be minimized, reducing potential bugs and inconsistencies.</p><h4><strong>11. Logs:</strong></h4><p>Treating logs as event streams provides valuable insights into the behavior and performance of microservices. Effective logging helps with debugging, monitoring, and analyzing system behavior, aiding in troubleshooting and performance optimization.</p><h4><strong>12. Admin Processes</strong></h4><p>Running administrative and management tasks as separate, one-off processes prevent interference with regular service operations. This approach ensures that administrative tasks can be executed without affecting the core functionality of microservices.</p><blockquote>For example, Python program using Virtualenv should use the vendored bin/python for running both the Tornado webserver and any manage.py admin processes.</blockquote><blockquote>By embracing these principles, organizations can unlock the full potential of microservices architecture and propel their applications to new heights of performance and scalability.</blockquote><p>Thanks for reading till the end, hope you found some insights from this.</p><p>Keep Learning &amp; Rising!</p><h4>Further Reading:</h4><ul><li><a href="https://gethelios.dev/blog/5-microservices-challenges-and-blindspots-for-developers/?utm_source=medium&amp;utm_medium=content&amp;utm_campaign=cloudnativedaily">5 Microservices Challenges and Blindspots for Developers</a></li><li><a href="https://gethelios.dev/blog/cut-engineering-costs-save-time-helios/?utm_source=medium&amp;utm_medium=content&amp;utm_campaign=cloudnativedaily">Microservices Monitoring: Cutting Engineering Costs and Saving Time</a></li><li><a href="https://python.plainenglish.io/top-11-tools-for-microservices-backend-development-in-2023-3d9cdd61ef10">Top 11 Tools for Microservices Backend Development in 2023</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=eb25e95e5906" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cloud-native-daily/what-are-12-factor-principles-of-cloud-native-microservices-eb25e95e5906">What Are the 12-Factor Principles of Cloud-Native Microservices?</a> was originally published in <a href="https://medium.com/cloud-native-daily">Cloud Native Daily</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to be more productive with help of a to-do list]]></title>
            <link>https://adityasonittyl.medium.com/how-to-be-more-productive-with-help-of-to-do-list-4ae6c90c2239?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/4ae6c90c2239</guid>
            <category><![CDATA[to-do-list]]></category>
            <category><![CDATA[hacks]]></category>
            <category><![CDATA[information-technology]]></category>
            <category><![CDATA[work]]></category>
            <category><![CDATA[productive]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Fri, 27 Jan 2023 14:10:56 GMT</pubDate>
            <atom:updated>2023-01-27T19:11:58.682Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aWRlEC9wXxQxhdtbTiEswA.jpeg" /></figure><p>Each and every human being have the same 24 hours a day, still, some builds system to serve millions of user, and yet another side struggle with time.</p><blockquote>It’s all about how you manage the time &amp; plan your day.</blockquote><p><em>A to-do list can make you more focused and productive, here’s how I make an effective to-do list before I start my day !!</em></p><p><strong>[0]</strong> 𝐔𝐬𝐞 𝐚 𝐟𝐨𝐫𝐦𝐚𝐭 that works for you to follow (it can be a physical notebook, digital tool, or a whiteboard)</p><p><strong>[1]</strong> Brainstorm and 𝐰𝐫𝐢𝐭𝐞 𝐝𝐨𝐰𝐧 𝐞𝐯𝐞𝐫𝐲𝐭𝐡𝐢𝐧𝐠 𝐧𝐞𝐞𝐝 𝐭𝐨 𝐝𝐨 to do or needs attention no matter big or small.</p><p><strong>[2]</strong> 𝐏𝐫𝐢𝐨𝐫𝐢𝐭𝐢𝐬𝐢𝐧𝐠 𝐭𝐚𝐬𝐤𝐬 which needs the most attention or importance. (I follow the Eisenhower matrix by putting tasks into four categories: urgent and important, important but not urgent, urgent but not important, and not urgent and not important.)</p><p><strong>[3]</strong> Break down tasks within 𝐬𝐮𝐛𝐭𝐚𝐬𝐤𝐬. (It will make you more progressive without feeling overwhelmed)</p><p><strong>[4]</strong> Setting up the 𝐫𝐢𝐠𝐡𝐭 𝐝𝐞𝐚𝐝𝐥𝐢𝐧𝐞𝐬. (Make non-disturbed slots of 25–30 mins, If something is taking time or you feel stuck take a break, and try again)</p><p><strong>[5]</strong> Review and 𝐫𝐞𝐟𝐥𝐞𝐜𝐭 𝐲𝐨𝐮𝐫 𝐭𝐨-𝐝𝐨 𝐥𝐢𝐬𝐭 𝐫𝐞𝐠𝐮𝐥𝐚𝐫𝐥𝐲 (Accomplished, need to do, new tasks)</p><p><strong>[6]</strong> Be 𝐟𝐥𝐞𝐱𝐢𝐛𝐥𝐞 𝐚𝐧𝐝 𝐜𝐡𝐚𝐧𝐠𝐞 𝐥𝐢𝐬𝐭 as needed (According to situations, adapt, update)</p><p>A to-do list is a living document, if you don’t feel able to complete something it’s okay, start it where you left off another day.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4ae6c90c2239" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[OpenShift Workloads from On-Premises to AWS Cloud [ROSA]]]></title>
            <link>https://adityasonittyl.medium.com/openshift-workloads-from-on-premises-to-aws-cloud-rosa-3706adabd32d?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/3706adabd32d</guid>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[openshift]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[kubernetes]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Thu, 25 Mar 2021 03:03:23 GMT</pubDate>
            <atom:updated>2021-03-25T04:22:17.135Z</atom:updated>
            <content:encoded><![CDATA[<p>If you are worrying about OpenShift workloads on On-Premises?, Or want to use a self managed service without having to spend time and resources managing the underlying platform and infrastructure, Then this article must be for you.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/715/1*5qT8JYDPV1tkxHIvyttjOA.png" /><figcaption>REDHAT-AWS</figcaption></figure><blockquote><em>𝗥𝗲𝗱 𝗛𝗮𝘁 𝗢𝗽𝗲𝗻𝗦𝗵𝗶𝗳𝘁 𝗦𝗲𝗿𝘃𝗶𝗰𝗲 𝗼𝗻 𝗔𝗪𝗦 (𝗥𝗢𝗦𝗔) is generally available now !</em></blockquote><p>So what’s the meaning of the above statement ? …. Let me simplify it for you.</p><h3><strong>What is ROSA ?</strong></h3><p>Red Hat &amp; Amazon Web Services (AWS) have collaborated to offer Red Hat OpenShift Service on AWS (ROSA), A fully managed and jointly supported offering, available via the AWS Management console. With this announcement, Now there is no need to worrying about your underlying infrastructure you can just focus on deploying your applications, with tons of workloads and Red Hat OpenShift will be the first third party offering available on the AWS console.</p><h3><strong>If you are already using OpenShift, How it can be helpful for you</strong> ?</h3><p><a href="https://www.openshift.com/products/amazon-openshift/">Red Hat OpenShift Service on AWS</a> Can simplify the ability to shift workloads to the cloud and making it even easier for you to adapt containers and deploy their applications faster on AWS, without having to spend time and resources managing the underlying platform and infrastructure.</p><h3><strong>What are the features it’s offering ?</strong></h3><ul><li>Access of ROSA on-demand from the AWS console.</li><li>Create fully-managed OpenShift clusters in minutes with self-service deployment, so you can get up and running quickly.</li><li>Receive the benefit of an integrated support &amp; billing experience with a single invoice from AWS and the ability to contact AWS or Red Hat for support.</li><li>Quickly build and deploy applications on Red Hat’s enterprise Kubernetes platform, using the same tools and APIs they would use on-premises or on any other public cloud.</li><li>Build containerized applications that integrate with the more than 170+ AWS cloud-native services.</li></ul><p>and much more on the way ….</p><h3><strong>If you was already using it on AWS Cloud, Now what’s the main advantage you can take from it?</strong></h3><p>Now you can accelerate your application development process by leveraging standard APIs and existing Red Hat OpenShift tools for your AWS deployments. You’ll also be able to take advantage and can integrate the wide range of AWS compute, database, analytics, machine learning, networking, mobile, and other services to improve customer agility, rate of innovation, and scalability.</p><h3><strong>How can one get started with creating and using the cluster ?</strong></h3><blockquote>Let’s start some hands-on to create a cluster and use it:</blockquote><ul><li>You must follow these <a href="https://access.redhat.com/documentation/en-us/red_hat_openshift_service_on_aws/4/html/setting_up_clusters_and_accounts/prerequisites">Prerequisites </a>and give appropriate permissions to start with below process.</li><li>Firstly you have to login in AWS and click on `Enable OpenShift` On ROSA service page.</li><li>By <a href="https://www.openshift.com/products/amazon-openshift/download">this </a>link you can download the ROSA-CLI and get started with creating the cluster.</li><li>Now you have to run below commands to get starting with it.</li></ul><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/994638befb955d746471ca5e94bfd74c/href">https://medium.com/media/994638befb955d746471ca5e94bfd74c/href</a></iframe><h3>References</h3><p><a href="https://www.openshift.com/blog/red-hat-openshift-service-on-aws-is-now-generally-available">Red Hat OpenShift Service on AWS is now Generally Available</a></p><p><a href="https://aws.amazon.com/blogs/aws/red-hat-openshift-service-on-aws-now-generally-availably/">Red Hat OpenShift Service on AWS Now GA | AWS News Blog (amazon.com)</a></p><p>That’s all for now … We will catch up again for some another uses cases.</p><p>Thanks for reading, If you have any queries we can sync at <a href="https://www.linkedin.com/in/theadisoni/"><strong>Linkedin</strong></a><strong>.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3706adabd32d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ArgoCD : GitOps Continuous Delivery Approach On Google Kubernetes Engine]]></title>
            <link>https://blog.searce.com/argocd-gitops-continuous-delivery-approach-on-google-kubernetes-engine-2a6b3f6813c0?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/2a6b3f6813c0</guid>
            <category><![CDATA[argo-cd]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[gitops]]></category>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[gke]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Fri, 20 Nov 2020 07:27:22 GMT</pubDate>
            <atom:updated>2024-05-22T07:50:21.208Z</atom:updated>
            <content:encoded><![CDATA[<h3>ArgoCD : GitOps Continuous Delivery Approach O<strong><em>n Google Kubernetes Engine</em></strong></h3><h4>If you are getting started with ArgoCD, Its Installation, Working, RollOuts, Deployments Strategies(BlueGreen, Canary) with Demo then you are landed on right page.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/811/1*cM-oD_QWXea-rYVEbtfFkA.jpeg" /><figcaption>ArgoCD : GitOps Continuous Delivery Approach On Google Kubernetes Engine</figcaption></figure><pre><a href="https://medium.com/@theadisoni/gitops-the-next-big-thing-for-devops-and-automation-2a9597e51559"><strong><em>Note : If you are new with GitOps approach then here&#39;s the previous part of this blog you can refer..!</em></strong></a></pre><h3>What is ArgoCD?</h3><p>ArgoCD is a tool made for GitOps continuous delivery in Kubernetes environments. Core components of ArgoCD is the Application Controller, That continuously monitors running applications state and compares that with live application state against the desired target state defined in the Git repository.</p><h3>Why ArgoCD?</h3><p>As we know GitOps strategy that rely on <strong><em>single source of truth </em></strong>that is <strong>Git,</strong> and when it comes to<strong><em> </em></strong>Application definitions, configurations, and environments should be declarative and version controlled where Application deployment and lifecycle management should be automated, auditable, and easy to understand all these we can achieve with help of ArgoCD.</p><h3>ArgoCD Architecture:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1002/1*3nPB4xER4aC2iXVs40K4Pw.jpeg" /><figcaption>ArgoCD Architecture</figcaption></figure><h3>Components:</h3><h4><strong>ArgoCD WebUI / CLI</strong></h4><p>Interaction component for preforming any task after installing ArgoCD on Cluster.</p><h4><strong>GitHub Repository</strong></h4><p>Repository contained all your Manifests-YAML files.</p><h4><strong>ArgoCD/API Server</strong></h4><p>The ArgoCD<strong>/</strong>API server is a gRPC/REST server which exposes the API consumed by the Web UI, CLI, and CI/CD systems. It has the following responsibilities:</p><ul><li>Manage the application and to report its status, Invoking of application operations (sync, rollback, user-defined actions)</li><li>Repository and cluster credential management (stored as K8s secrets), Authentication and auth delegation to external identity providers, Role-based access control (RBAC) enforcement</li><li>listener/forwarder for Git webhook events.</li></ul><h4><strong>Repository Server</strong></h4><p>Internal service that maintains local caches of the Git repo with the application manifests. it is mainly responsible for generating and returning the Kubernetes manifests when provided the following inputs:</p><ul><li>Repository URL, revision (commit, tag, branch), application path</li><li>Template-specific settings: ksonnet environments, parameters, helm values.yaml.</li></ul><h4><strong>Application Controller</strong></h4><p>A<strong> </strong>Kubernetes controller which continuously monitors running applications and compares the current, live state against the desired target state (as specified in the repo). It detects <em>OutofSync</em> application states and optionally takes the corrective actions. This mainly responsible for invoking any user-defined hooks for lifecycle-events ( Sync, PostSync, PreSync)</p><h4><strong>Redis &amp; Dex-Server</strong></h4><p>Redis is only used as a throw-away cache and can be lost and rebuilt without loss of service. The Dex-Server uses an in-memory database, and two or more instances would have inconsistent data.</p><h4><strong>ArgoCD App</strong></h4><p>Isolated environment for a particular application running via a repo.</p><h4><strong>Kubernetes Clusters</strong></h4><p>You can select where you want to run application on default cluster or any other one, Dex-Server will trigger that for you!.</p><h3>How does ArgoCD work?</h3><p>ArgoCD is implemented as Kubernetes CRD<strong> </strong>which continuously monitors running applications with compares the current, live state against the desired target states (as per git repo). The deployed application whose live state deviates it from its target state is considered <em>OutofSync</em>.</p><p>ArgoCD reports &amp; visualizes any deviation and provides mechanisms to automatically or manually sync the live and desired target state. Any modifications made to the desired target state in the git repo can be automatically applied and reflected in the specified target environments.</p><p>You can get started in 2 easy steps:</p><ul><li>Define your App and environment configs in a Git repo</li><li>Argo monitors and syncs your app between Git config and the Kubernetes cluster using any pipeline (Argo CI, Jenkins, etc.), CLI or UI.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bYYcHmc1fHev_DBfs27dAA.png" /><figcaption>How does ArgoCD work?</figcaption></figure><p>ArgoCD supports both the pull and the push-based GitOps model to sync target environments with desired application state. For a customer we helped recently, we followed the push model and invoked ArgoCD from a CD pipeline (Jenkins) for security and compliance reasons.</p><h3>Install and set up the ArgoCD:</h3><p><strong>Prerequisites: </strong>Before getting started you will need a running GKE Cluster or Kubectl tool installed with <a href="https://kubernetes.io/docs/tasks/access-application-cluster/configure-access-multiple-clusters/">kubeconfig</a> (default location is ~/.kube/config).</p><h4><strong>1. Install ArgoCD</strong></h4><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/0c0d5551a482d5ea7ec945adf1c65755/href">https://medium.com/media/0c0d5551a482d5ea7ec945adf1c65755/href</a></iframe><p>This will create a new namespace argocd where ArgoCD services and application resources will live, If you want configurations as per your use case then edit given yaml then apply.</p><h4><strong>2. Install ArgoCD CLI</strong></h4><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/29332fe94377a3771123c4202ce8718f/href">https://medium.com/media/29332fe94377a3771123c4202ce8718f/href</a></iframe><blockquote>for windows try <a href="https://argoproj.github.io/argo-cd/cli_installation/#windows">this</a></blockquote><h4><strong>3. Expose ArgoCD API Server</strong></h4><p><strong>(a)</strong> Change the argocd-server service type to LoadBalancer:</p><pre>kubectl patch svc argocd-server -n argocd -p ‘{“spec”: {“type”: “LoadBalancer”}}’</pre><p><strong>(b) </strong>Ingress / Port Forwarding:</p><p>Follow the <a href="https://argoproj.github.io/argo-cd/operator-manual/ingress/">ingress documentation</a> on how to configure ArgoCD with ingress. Kubectl port-forwarding used for connect to the API server without exposing the service.</p><pre>kubectl port-forward svc/argocd-server -n argocd 8080:443</pre><p>After it you can access API server at localhost:8080</p><h4><strong>4. Login</strong></h4><p><strong>(a) Using The CLI</strong></p><p>The initial password is auto-generated to be the pod name of the ArgoCD API server. This can be retrieved with the command:</p><pre>kubectl get pods -n argocd -l app.kubernetes.io/name=argocd-server -o name | cut -d&#39;/&#39; -f 2</pre><p>The above command will give you initial password use with username admin login to ArgoCD&#39;s IP or hostname:</p><pre>argocd login &lt;ARGOCD_SERVER&gt; <br>argocd account update-password #Change the password using this</pre><p><strong>(b) Login Using GUI</strong></p><p>After expose check ip and hit from browser leave SSL warning and put user as “admin” and password got from above cmd.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*27brqf9aWLZjF3vnv16fdg.png" /><figcaption>Exposed IP in Namespace</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/305/0*CtUIXpGuDbKYb2ih.png" /></figure><h4><strong>5. Creating Apps Via UI</strong></h4><p>After logging in, click the “<strong>+ New App”</strong> button as shown.</p><p>and fill details like:</p><p>Give any name to app, in the project default, and leave the sync policy as Manual.</p><blockquote>Set repository url to the github repo url, leave revision as HEAD, and set the path to “your_folder_at_repo”.</blockquote><blockquote>For <strong>Destination</strong>, set cluster to in-cluster and namespace to default. (or set to another cluster config you want)</blockquote><blockquote>After filling out the information above, click <strong>Create</strong> at the top of the UI to create the “example” application</blockquote><h4><strong>6. Creating Apps Via CLI</strong></h4><p>To Create application via CLI here’s included example syntax you can modify name, project, repoURL, path, dest, as per your use-case and apply it.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/3b621b78eba8ea58547000df8adf336c/href">https://medium.com/media/3b621b78eba8ea58547000df8adf336c/href</a></iframe><pre>kubectl apply -f app.yaml<br>#application.argoproj.io/example-app created<br>argocd app get example-app<br>#info of example-app</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/328/1*ck6dz41i5kU_b30Qc39Ywg.png" /></figure><p>After applying at GUI you can see there a app is deploying which is now “OutOfSync” and “Syncing” at this stage it is deploying given manifests and recording stage of app with current version. you can fetch same app in your namespace via kubectl like:</p><pre>kubectl get all -n argocd </pre><p>After click on above application you will see output like below image or ArgoCD App deployed each deployments, svc, pods.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PvnQFnRO8Hm2TINFbY1aRw.png" /><figcaption>Deployed application</figcaption></figure><p>For now you can try by simple commit at Git and sync here all changes will be reflected shortly!</p><h3><strong>RollOuts in ArgoCD</strong></h3><p>K8s Deployments provides the RollingUpdate strategy with a basic set of safety guarantees during updates. However the rolling update strategy faces many limitations For overcome those reasons, in large scale high-volume production environments we move to ArgoCD Strategies with deployment capabilities such as experimentation , blue-green, canary, progressive delivery and canary analysis features to Kubernetes.</p><p>Argo Rollouts (optionally) integrates with ingress controllers and service meshes, leveraging their traffic shaping abilities to gradually shift traffic to the new version during an update.</p><h3><strong>Installation: Argo Rollouts and Kubectl CLI Plugin</strong></h3><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/2abf306b0e3d0ec89fa85401068f4135/href">https://medium.com/media/2abf306b0e3d0ec89fa85401068f4135/href</a></iframe><p>These three Points to be noted while writing Manifests from a deployment to Rollout:</p><ol><li>Changing kind from Deployment to Rollout</li><li>Changing apiVersion to argoproj.io/v1alpha1</li><li>Adding a specific deployment strategy in the rollout (Blue Green in the example below)</li></ol><p>ArgoCD Rollouts currently Supports Blue Green and ReplicaSet Based Canary.</p><h3><strong>Deployment Strategies in ArgoCD</strong></h3><h4><strong>Blue-Green</strong></h4><p>When deploying a new version of an application, a second environment is created. Once the new environment is tested, it takes over from the old version, and you can off the old environment.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/95f6977600bfc14d5054b8f9c93d7e14/href">https://medium.com/media/95f6977600bfc14d5054b8f9c93d7e14/href</a></iframe><p>After applying any changes in manifests you will see bluegreen deployment as below image on same example-app we used before, for load sharing it started creating same instance of new configuration and shifting load to them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/721/1*1Qj9lBxijUTKHkHM1okyWA.png" /><figcaption>Blue-Green working</figcaption></figure><h4><strong>Canary</strong></h4><ul><li>A new version of a micro-service is started along with the old versions. New version can then take a portion of the requests and the team can test how this new version interacts with the overall system.</li></ul><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/1c670852c3a1cb6d9d81396394873b62/href">https://medium.com/media/1c670852c3a1cb6d9d81396394873b62/href</a></iframe><p>After applying any changes in manifests you will see canary deployment as below image on same example-app we used before, Load is started to share for v2 version from rset.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/905/1*Fw9sA3EoWyxjEIIWddoejw.png" /><figcaption>Canary working</figcaption></figure><p>That’s all for this time here’s some references that can help!</p><h3>References</h3><ul><li><a href="https://argoproj.github.io/argo-cd/">Argo CD - Declarative GitOps CD for Kubernetes</a></li><li><a href="https://github.com/argoproj">Argo Project</a></li></ul><blockquote>I hope It was helpful. If you enjoyed this article, feel free to clap and share it with a friend. If you want to read about many more such amazing use-cases and applications, Stay tuned to <a href="https://medium.com/searce">Searce Engineering</a>, for any queries or suggestions here’s my <a href="https://www.linkedin.com/in/theadisoni/">LinkedIn profile</a>!</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2a6b3f6813c0" width="1" height="1" alt=""><hr><p><a href="https://blog.searce.com/argocd-gitops-continuous-delivery-approach-on-google-kubernetes-engine-2a6b3f6813c0">ArgoCD : GitOps Continuous Delivery Approach On Google Kubernetes Engine</a> was originally published in <a href="https://blog.searce.com/">Searce</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[GitOps : The Next Big Thing for DevOps and Automation!]]></title>
            <link>https://blog.searce.com/gitops-the-next-big-thing-for-devops-and-automation-2a9597e51559?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/2a9597e51559</guid>
            <category><![CDATA[automation]]></category>
            <category><![CDATA[deployment]]></category>
            <category><![CDATA[gitops]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[linux]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Fri, 20 Nov 2020 07:26:05 GMT</pubDate>
            <atom:updated>2020-11-20T07:26:05.895Z</atom:updated>
            <content:encoded><![CDATA[<h3>GitOps : The Next Big Thing for DevOps and Automation!</h3><h4>If you have similar questions like: “What’s GitOps?”, “Why we are moving towards this?”, “How and when one can implement this strategy in now running environment?”, “What are the tools it included?” then you have landed on the right page.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/907/1*CsXJ8dt3aFaXan3LJn-jxA.png" /><figcaption>GitOps : The Next Big Thing for DevOps and Automation!</figcaption></figure><p>You must be familiar with the terms DevOps, DevSecOps, and even NoOps. But if you haven’t already, you’re going to be hearing a whole lot more about GitOps.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/400/1*PZLPS-m1_iGNudZPyfhewA.gif" /><figcaption>Can’t Wait more to get started..</figcaption></figure><blockquote><strong>So let’s start..</strong></blockquote><h3>From where it’s all started?</h3><p>In 2017 Weaveworks, which specializes in Kubernetes solutions, cameup with the idea of ‘GitOps’, and now GitOps has become the latest trend in Cloud Native development.</p><h3>What is GitOps?</h3><p>GitOps is the tradition of using Git pull requests to verify and automatically deploy system infrastructure modifications with an automated strategy. GitOps comes with main idea to shift focus from the infrastructure to the Git repo basically, means when you want to deploy a any new application or configure a recent one. You should <strong>rely on Git itself, instead of servers</strong>, “Just prepare the changes”, “Create pull requests”, “pass the reviews process and”, “merge”, That’s how GitOps magic happens!.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/747/1*fbf4rvYnqoGSUpCXh32-iA.png" /><figcaption>Credits: <a href="https://github.com/weaveworks/awesome-gitops">https://github.com/weaveworks/awesome-gitops</a></figcaption></figure><p>It treats Git as the <strong><em>single source of truth</em></strong>, for controlling mechanism for system architecture(Include: creating, updating, and deleting).</p><p>GitOps based on a Git-based source code management system and hence GitLab, GitHub, and Bitbucket are same choices. Bitbucket also given a GitOps-y approach named <a href="https://developer.atlassian.com/blog/2017/07/kubernetes-workflow/">BDDA</a>, pronounced like ‘Buddha’, for “Build-Diff, Deploy-Apply”.</p><h3>Why we are moving towards this?</h3><p>See in today’s world automation is a key for CI/CD Automated delivery pipelines push changes made to Git into your infrastructure. GitOps goes a step further. It uses tools, like Flux, ArgoCD, etc to automate the comparison between the state of a cluster and the Git configuration.</p><p>GitOps by operator in cluster triggers the deployments for Kubernetes, and make need for other CD tools more redundant. so it’s already a reliable approach to move on with.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qdmteM01fDFCMD8lKs0l4g.png" /><figcaption>GitOps Pipeline Overview</figcaption></figure><p>That implies CI doesn’t need access to the cluster. Each update automatic transit for Git logs. Transactions either succeed in a clean state or fail, That’s how via choosing approach of code-based we don’t need new infrastructure each time.</p><h3>Benefits of the GitOps approach :</h3><ul><li>Expanded Productivity</li><li>Progressed Developer Experience</li><li>Auditing</li><li>More powerful Reliability</li><li>Elasticity and Uniformity</li><li>Stable Security Guarantees</li></ul><h3>How GitOps Works?</h3><blockquote>GitOps Works based upon pull pipelines</blockquote><p><strong>Pull pipelines <em>: </em></strong>Assume you have every separate part of the documentation, all of your YAML files, and every bit of necessary code required for Kubernetes included in a single Git repository. by help of a some automation tools, anyone can do a pull request, edit the code, and then issue a merge request to that Git repository.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1021/1*hHS0rQ4FIzgS5hlz_OZ3gw.png" /><figcaption>Pull Based Pipeline</figcaption></figure><p>Once the merge request is complete, the automated GitOps operator detects something has changed, instead of using push-based we use here the pull-based approach means A pull pipeline involves a K8s Operator(Flux) and deploying new images from inside the cluster. <br>It detects any new images in the Git repository and pulls them into the cluster environment. This bypasses cluster credentials being exposed outside of the production environment.</p><h4>Pull vs Push Pipeline</h4><p><strong>Push Pipelines : </strong>As<strong> </strong>today lots of CI/CD tools uses model of push-based for deployments. A push-based pipeline means that code starts with the CI system and may continue its path through a series of encoded scripts or uses ‘kubectl’ by hand to push any changes to the Kubernetes cluster.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RwmQaVD2BipOBLo1zEvhBA.png" /><figcaption>Push Based Pipelines</figcaption></figure><p>The reason you don’t want to use your CI system as the deployment reason or do it manually on the command line is because of the potential to expose credentials outside of your cluster. while it is already possible to secure both CI/CD scripts and the CLI , this is generally not good practice and is why CI systems can be known as attack vectors for production.</p><h3>What are the tools it included?</h3><blockquote><strong><em>GitOps Is for the entire Cloud Native - Terraform &amp; More..</em></strong></blockquote><p>GitOps is not only about the K8s but also if we want the whole system to be managed declaratively and applying convergence by a “whole system,” we involve a group of environments running Kubernetes, e.g. “dev-cluster,” “prod-cluster.” Each environment includes machines, clusters, applications, as well as interfaces to external services exp: data, monitoring. On the same scenario we can notice how Terraform can work for the bootstrapping problem here. Via GitOps workflows with K8s and IAC(Terraform) we can create single control plane that underpins Kubernetes and applications via Git as “Single source of truth”. This is a helpful most beneficial practice.</p><h3>Tools</h3><ul><li><a href="https://github.com/argoproj/argo-cd">ArgoCD</a> — Declarative continuous delivery for Kubernetes</li><li><a href="https://www.runatlantis.io/">Atlantis</a> — Terraform pull request automation</li><li><a href="https://github.com/autoapply/autoapply">Autoapply</a> — Automatically apply changes from a Git repository to a Kubernetes cluster</li><li><a href="https://rollout.io/">CloudBees Rollout</a> — Feature Flag as-a-Service that leverages GitOps &amp; Config-as-Code (commercial product from CloudBees)</li><li><a href="https://github.com/fluxcd/flux">FluxCD</a> — The GitOps Kubernetes operator</li><li><a href="https://github.com/fluxcd/helm-operator">Helm Operator</a> — Automates Helm Chart releases in a GitOps manner</li><li><a href="https://github.com/weaveworks/flagger">Flagger</a> — Progressive delivery Kubernetes operator (Canary, A/B testing and Blue/Green deployments automation)</li><li><a href="https://github.com/weaveworks/ignite">Ignite</a> — A Virtual Machine manager with a container UX and built-in GitOps</li><li><a href="https://github.com/pusher/faros">Faros</a> — CRD based GitOps controller</li><li><a href="https://gitkube.sh/">Gitkube</a> — Build and deploy docker images to Kubernetes using git push</li><li><a href="https://jenkins-x.io/">Jenkins X</a> — a CI/CD platform for Kubernetes that provides pipeline automation, built-in GitOps and preview environments</li><li><a href="https://www.kubestack.com/">KubeStack</a> — GitOps framework using Terraform for Cloud Kubernetes distros (AKS, GKE, and EKS) with CI/CD examples for common tools</li><li><a href="https://www.weave.works/product/cloud/">Weave Cloud</a> — GitOps experience, workflows and dashboard for your cluster(s) (commercial product from Weaveworks)</li><li><a href="https://werf.io/">Werf</a> — GitOps tool with advanced features to build images and deploy them to Kubernetes (integrates with any existing CI system)</li><li><a href="https://pipecd.dev/">PipeCD</a> — Continuous Delivery for Declarative Kubernetes, Serverless and Infrastructure Applications.</li></ul><h3>References</h3><ul><li><a href="https://www.weave.works/technologies/gitops/">GitOps what you need to know</a></li><li><a href="https://github.com/weaveworks/awesome-gitops">GitHub - weaveworks/awesome-gitops: A curated list for awesome GitOps resources</a></li></ul><p><a href="https://medium.com/@theadisoni/argocd-gitops-continuous-delivery-approach-on-google-kubernetes-engine-2a6b3f6813c0"><strong><em>Here’s next blog link for getting started with ArgoCD Deployments for kubernetes on GKE.</em></strong></a></p><blockquote>I hope It was helpful. If you enjoyed this article, feel free to clap and share it with a friend. If you want to read about many more such amazing use-cases and applications, Stay tuned to <a href="https://medium.com/searce">Searce Engineering</a>, for any queries or suggestions here’s my <a href="https://www.linkedin.com/in/theadisoni/">LinkedIn profile</a>!</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2a9597e51559" width="1" height="1" alt=""><hr><p><a href="https://blog.searce.com/gitops-the-next-big-thing-for-devops-and-automation-2a9597e51559">GitOps : The Next Big Thing for DevOps and Automation!</a> was originally published in <a href="https://blog.searce.com/">Searce</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[MLOps: The Epoch of Productionizing ML Models]]></title>
            <link>https://medium.com/analytics-vidhya/mlops-the-epoch-of-productionizing-ml-models-4eec06d93623?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/4eec06d93623</guid>
            <category><![CDATA[docker]]></category>
            <category><![CDATA[cicd-pipeline]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[linux]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Thu, 11 Jun 2020 04:37:38 GMT</pubDate>
            <atom:updated>2020-06-19T16:03:43.694Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/863/1*b84WiZOp8DTQVrNd1kjGIA.png" /></figure><p>If You have Similar questions like “<strong>What is MLOps?</strong>”, “<em>Why One Should adopt MLOps?</em>”, “<strong>DevOps vs MLOps</strong> What are Differences and when to use which?”, “<strong>How to Adopt MLOps</strong>?”.</p><blockquote>Then you are landed at the right place.</blockquote><p><strong>This Blog I am breaking into two parts</strong></p><p><a href="https://medium.com/@theadisoni/mlops-the-epoch-of-productionizing-ml-models-4eec06d93623"><strong>Part 1: </strong>Originate with MLOps.</a></p><p><a href="https://medium.com/@theadisoni/mlops-in-azure-f6e7c006fe0e"><strong>Part 2: </strong>MLOps in Azure.</a></p><h4>So let’s Start with Part 1: <strong>Originate with MLOps</strong>.</h4><p><strong>MLOps</strong> (a compound of Machine Learning and “information technology OPerationS”) is new discipline/focus/practice for collaboration and communication between <strong>data scientists and information technology</strong> (IT) professionals while <strong>automating </strong>and <strong>productizing </strong>machine learning algorithms.</p><h4><strong>What is MLOps?</strong></h4><p>In simple terms, <strong>MLOps</strong>(DevOps for ML) is the Machine learning equivalent of DevOps. While DevOps helped optimize the production lifecycle of Big Data project, <strong>MLOps</strong> seeks to solve the problems associated with the implementation of <strong>ML in production.</strong></p><h4><strong>From DevOps to MLOps:</strong></h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*9gbbQrdMD94-yZpXswUAfw.jpeg" /></figure><h4>Why adopt MLOps?</h4><p>For businesses implementing ML, <strong>MLOps</strong> should be a top priority. It can streamline business processes in the following ways:</p><ul><li><strong>MLOps</strong> can take care of regulatory and compliance issues. While it introduces the Operations team to the latest regulations and best industry practices, it allows the Data Science team to focus on building and deploying creative ML models.</li><li><strong>Machine Learning Operalization</strong> helps reduce the gap between discovering insights and turning those insights into actionable business plans.</li><li>By allowing <strong>Data Scientists and Operations teams</strong> to work together, <strong>MLOps</strong> combines the power of business knowledge with data. This encourages data-driven business operations.</li></ul><h4>How is MLOps different from DevOps?</h4><ul><li>Data/model versioning <strong>!=</strong> code versioning</li><li>Retraining ability requires on-demand as the models decay over time.</li><li>Model reuse entirely has different case than <strong>software </strong>reuse, as models need tuning based on <strong>scenarios </strong>and data.</li><li>Fine-tuning is needed when to reuse a model. Transfer learning on it, and it leads to a <strong>training pipeline.</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/879/1*QwujZsLJjWI5eTNOPr6Drg.png" /></figure><h4>How to Adopt MLOps?</h4><p>For <strong><em>fault tolerance</em></strong> and iterative filtration of <strong><em>ML models is essential</em></strong>, and for that reproducibility is needed. Repeatability required to <strong>illuminate the source</strong> of variation like –</p><ul><li>CPU<strong> multi-threading</strong></li><li>Non-deterministic<strong> GPU-floating</strong> point calculation</li><li>Inconsistent<strong> hyper parameters</strong></li><li><strong>Shuffling </strong>of datasets</li><li>Noisy hidden layers</li><li>Change to <strong>model </strong>architecture</li><li>Random initialization of layers weights</li><li>Change in <strong>ML frameworks</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/848/1*jJK19QuWj08ultCSP_ncHw.png" /></figure><h3>ML Operations:</h3><h4>Controllability</h4><p>When it proceeds to establish the <strong>production updates </strong>it’s really difficult in <strong>ML pipelines</strong> as there is not only the origin code that changes in the pipeline but when a new retrained model is selected by<strong> human approval </strong>or some advance auto-selection method, The new changes should be done with proper control of not having any instability or downtime of <strong>ML applications</strong>.</p><h4>Automation</h4><p><strong><em>ML pipelines are code, and DevOps toolchain pipeline plays an essential role in MLOps.</em></strong> The source code repository automation facilitated by Jenkins and orchestrators such as <a href="https://airflow.apache.org/">AirFlow</a> is a classic example. But when it comes to ML pipelines, There are interdependencies like <strong>Model Approval</strong> and Drift Detection. These additional dependencies need to be integrated with the <strong>DevOps toolchain pipeline.</strong></p><h4>Model Supervision</h4><p>Model Management is the <strong>core of MLOps</strong>. When it comes to managing complex pipelines where a large number of models, objects, and <strong>training pipelines</strong> are generated MLOps needed.</p><h4>Model versioning</h4><p>Making changes reversibly is necessary for production for stability and fault tolerance. Unlike source code versioning of<strong> ML models</strong> is an additional step from traditional pipeline.</p><h4>Model tracking</h4><p><strong>Complex pipelines </strong>of models lead to many models runs in the pipeline like in case of ensembles models runs. Creating this many experiments to select the best champion or challenger model requires model tracking. There are different kinds of tools like <strong>MLflow </strong>available for tracking models, or one can build a custom pipeline according to the use case.</p><h4>ML Monitoring</h4><p><strong><em>Monitoring </em></strong>in ML systems is not just checking the uptime of different services and resources/compute its taking. As in <strong>production</strong>, ML brings things to monitor that is directly related to the success of <strong>business outcomes.</strong></p><h4>ML data drift</h4><p>The change in the relationship between <strong>input and output </strong>data with time called data to drift. The data drift analysis should be in the ML monitoring. The observability of data drift is essential to analyze if retraining requires any changes in the configuration of ML models. That is the reason why inference monitoring is needed. Let’s discuss <strong>inference monitoring</strong> in brief.</p><h4>Inference Monitoring</h4><p><strong>Continuously monitor</strong> the inference and observe if it is behaving according to the expected bounds. This <strong>monitoring tell</strong>s about mismatching of input and output data which detects data drift, and it also provides the performance of ML model using which performance analysis and <strong>comparison</strong> of the models is possible.</p><h3>Benefits of Enabling MLOps:</h3><ul><li>It provides the advantage to strap and increase investments in existing machine learning and data science tools and technology to the maximum and create a consolidates, a system of maintaining the records in between <strong>different teams and different projects</strong></li><li>It helps to manage and maintain the <strong>partnership of data science and IT/Ops teams to work together</strong> to provide ML-powered applications which can provide some values in results.</li><li>It helps to<strong> minimize the hazard to the organization</strong> by regularizing and importing in place hefty governance checklist and balances, and by enabling the use of <strong>best practices for machine learning models</strong> at the production level.</li><li><strong>Scale ML Models initiatives broadly</strong> by swiftly and flawlessly converting the true potential of machine learning into business processes (which already exist) and <strong>systems </strong>(which exist across the enterprise)</li></ul><blockquote>That’s all about this part in <a href="https://medium.com/@theadisoni/mlops-in-azure-f6e7c006fe0e">part 2: MLOps in Azure</a></blockquote><ul><li>Introduction: <a href="https://medium.com/@theadisoni/mlops-in-azure-f6e7c006fe0e">MLOps in Azure</a></li><li>Overview</li><li>Azure MLOps features</li><li><a href="https://medium.com/@theadisoni/mlops-in-azure-f6e7c006fe0e">Create a project in Azure DevOps with CI/CD for MLOps</a></li></ul><p>Lets catch up at <a href="https://medium.com/@theadisoni/mlops-in-azure-f6e7c006fe0e">part 2</a></p><blockquote><strong>Thank You</strong></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4eec06d93623" width="1" height="1" alt=""><hr><p><a href="https://medium.com/analytics-vidhya/mlops-the-epoch-of-productionizing-ml-models-4eec06d93623">MLOps: The Epoch of Productionizing ML Models</a> was originally published in <a href="https://medium.com/analytics-vidhya">Analytics Vidhya</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[MLOps In Azure…]]></title>
            <link>https://adityasonittyl.medium.com/mlops-in-azure-f6e7c006fe0e?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/f6e7c006fe0e</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[mlops]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Thu, 11 Jun 2020 04:37:10 GMT</pubDate>
            <atom:updated>2020-06-11T04:42:00.285Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*MApNoN9IRaGqC_Ei.jpg" /></figure><p>In <a href="https://medium.com/@theadisoni/mlops-the-epoch-of-productionizing-ml-models-4eec06d93623">Part 1</a> Of this Blog We already discussed ….<a href="https://medium.com/@theadisoni/mlops-the-epoch-of-productionizing-ml-models-4eec06d93623">What is MLOps ?</a> <a href="https://medium.com/@theadisoni/mlops-the-epoch-of-productionizing-ml-models-4eec06d93623">Why MLOps ?</a> <a href="https://medium.com/@theadisoni/mlops-the-epoch-of-productionizing-ml-models-4eec06d93623">DevOps vs MLOps?..</a></p><p>Now its time to do some hands-on with MLOps in Azure.</p><p>This series of posts will assume the reader has some knowledge &amp; experience of <a href="https://docs.microsoft.com/en-us/azure/machine-learning/team-data-science-process/overview">data science workflows</a>.</p><p>In this blog we’ll set up pipelines to create a minimal end-to-end MLOps pipelines to achieve the following using Azure Machine Learning and Azure Pipelines:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*isfytmBNNCeTCFSb.png" /></figure><p>Across this series of posts, we will create 5 Azure Pipelines in this example project.</p><p>Each of these pipelines are on an automated trigger — triggered either by a change to code or on a schedule. However, if we make changes and want to trigger the pipelines manually, we can do so through Azure Pipelines.</p><p><em>The pipelines are as follows:</em></p><h4>Data Pipeline</h4><p>First we’ll have a data Pipeline to create a dataset and upload it to Azure Blob Storage. This datastore will then be registered with Azure Machine Learning ready for using in our model training pipeline.<em>We’ll set this up as a daily pipeline.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DwKiIMvyenIgrFvg.png" /></figure><h4>Environment Pipeline</h4><p>Second, we’ll create an Azure ML Environment using a custom python package. The custom python package we’ll use is in our Azure Repos git repository.</p><p>Often we’ll have one or more internal packages that will be used for sharing code for repeatable tasks such as custom data transformation we’d like to share across model training and model scoring.</p><p>We also install any external packages we need into this environment.</p><p>We’ll want to use this python package in our model training and model deployment, so each time we merge our code into our master branch, we’ll update the Azure ML Environment with our custom python package.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*w8T5E9-PyWut4zVN.png" /></figure><h4>Model Training Pipeline</h4><p>The third pipeline we’ll create is a model training Pipeline to train our ML model and register it to Azure ML Models.</p><p>In this pipeline we set up the compute node we’ll be using for training and, on this compute node, we pull in the environment we set up in the previous pipeline. We also mount the datastore we registered in our data pipeline for training our model.</p><p>We’ll have this pipeline on a schedule to retrain the model once per week.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*XEkbhAyS3VHqKF_R.png" /></figure><h4>Model Deployment Pipeline</h4><p>Once we’ve got a trained, registered model, we’ll have a Model Deployment Pipeline to deploy the trained models to a web service using Azure Container Instances.</p><p>We will also run this once a week once our model training/retraining pipeline is complete.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*wMj370_2CgekjpzC.png" /></figure><h4>CI Pipeline</h4><p>The other pipelines are focused on the tasks that you’ll probably be more interested in as a data scientist. However, it’s important to note that you’ll often be working on code with others and this will be production code — so it’s important to enforce a set of coding standards and a suite of unit tests to run.</p><p>In our CI pipeline, every time a user makes a pull request in git, we’ll be running pylint for static code analysis and for automated running of tests. We’ll publish the test results and code coverage back to Azure DevOps for users to view before choosing to merge the code.</p><p>The following CI pipeline is the one I often use:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*zzAfDUWV18mGMBEq.png" /></figure><h3>Why Azure Pipelines?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*2iJg8zByaykdlp0F.png" /></figure><p>We’ll be using Azure Pipelines in this example to set up and trigger pipelines.</p><p>Azure Pipelines are cloud-hosted pipelines that are fully integrated with Azure DevOps. You can either use a yaml file or a UI-based tool in Azure DevOps to set up your pipelines.</p><p>They can be used with code stored in a range of repository locations, including Azure Repos and Github.</p><p>I’m personally a fan of Infrastructure as Code (IaC) so we’ll be using the yaml definitions of pipelines. As taken straight from the zen of python — “Explicit is better than implicit” — and I think you should only have to look at your code to know what will be deployed and run during these pipelines.</p><p>Azure Pipelines include a huge range of functions for running pipelines for tasks such as testing, static code analysis, building, deploying etc.</p><p>As an alternative, we could use Azure Data Factory, which includes a full assortment of integrated ETL tools, to manage the pipelines but as we won’t need any of this additional functionality and to keep resources to a minimum we’ll stick to Azure Pipelines.</p><h3>Why Azure Machine Learning?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ggNUEQqDtZ3meSDE.png" /></figure><p>Azure Machine Learning provides a whole host of functionality to accelerate the end-to-end machine learning lifecycle to help deploy enterprise-ready ML solutions quickly and robustly.</p><p>We’ll be using Azure Machine Learning for a range of tasks concerning the management of the ML lifecycle including:</p><ul><li>Registration of data sources and mounting of data storage to training environment</li><li>Storage and versioning of the environment in which to execute training and scoring</li><li>Managing compute for training experiments</li><li>Tracking training experiment runs and associated metrics</li><li>Storage and versioning of trained ML models</li><li>Deployment of models to a web service using Azure Container Instances (Azure Kuberenetes Service also available)</li></ul><h3>Set Up Azure DevOps</h3><p>Navigate to <a href="https://dev.azure.com/">dev.azure.com</a> to sign into Azure DevOps.</p><p>You’ll be greeted with a page similar to the one below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BQloDUysd2Z8CvpN.png" /></figure><p>Click on the “+ New Project” Button. Fill in the name of your project and description and click Create:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/634/0*riH65xHapbQFOhKR.png" /></figure><p>Now navigate to Azure Repos within your new DevOps project, you should be greeted with a page that looks similar to the screenshot below. Click on the repo dropdown at the top of the page and click “Import Repository” to import the <a href="https://dev.azure.com/bekeen/_git/MLOps-Example-Project">git repository that accompanies these posts</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5geZ8dJKH0V93U3N.png" /></figure><p>The clone URL you’ll need to import the repository is <a href="https://dev.azure.com/bekeen/MLOps-Example-Project/_git/MLOps-Example-Project">https://dev.azure.com/bekeen/MLOps-Example-Project/_git/MLOps-Example-Project</a></p><p>Once imported, you’ll have the git repository in your Azure Repos within your Azure DevOps project, as below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*dPcmSt6uZoIQHpHv.png" /></figure><h3>Create Azure Resources using Azure CLI</h3><p>With our Azure DevOps Project set up, we can set up the rest of our resources using the Azure CLI.</p><h3>Create a Resource Group</h3><p>The first thing we’ll need to do is create a resource group in which to store all of our resources.</p><p>Using the Azure CLI tool run:</p><pre>az group create --name &lt;resource-group&gt; --location &lt;location&gt;</pre><h3>Create an Azure Machine Learning Workspace</h3><p>Next we’ll create an Azure Machine Learning Workspace, this workspace will manage:</p><ul><li>Environments that can be used for training and scoring</li><li>We will be using a custom python package that we will be defining in this project.</li><li>This package will be pip installed into an Azure ML Environment.</li><li>We will manage our environment using the Azure ML Workspace.</li><li>Models</li><li>Trained models will be stored in our Workspace.</li><li>Our scoring script can then retrieve the latest models.</li><li>Datastores in which our training data is stored</li><li>We will be using a static dataset that we will create here.</li><li>This could, however, be an ever changing dataset in blob storage, Azure Datalake etc. from which retraining could be done.</li></ul><p>Using the Azure CLI tool run:</p><pre>az ml workspace create -w &lt;workspace-name&gt; -g &lt;resource-group&gt;</pre><h3>Create an Azure Storage Account</h3><p>The Azure storage account will be used to store a train and test dataset, as well as a validation dataset</p><p>Using the Azure CLI tool run:</p><pre>az storage account create --name &lt;storage-account-name&gt; <strong>\</strong><br>    --resource-group &lt;resource-group&gt; <strong>\</strong><br>    --location &lt;location&gt; <strong>\</strong><br>    --sku Standard_ZRS <strong>\</strong><br>    --encryption blob</pre><p>We’ll need the key for this storage account so that we can store data to and retrieve data from this storage account. To get this key, we can run using the Azure CLI tool:</p><pre>az storage account keys list --account-name &lt;storage-account-name&gt; --resource-group &lt;resource-group&gt;</pre><h3>Create a Service Principal with Password Authentication</h3><p>Our service principal credentials can be used to authenticate access to our Azure ML Workspace.</p><p>Using the Azure CLI tool run:</p><pre>az ad sp create-for-rbac --name &lt;spn-name&gt;</pre><p><em>It is important to note down the app id and password from the creation of this service principal! You’ll also need the tenant ID listed here</em></p><h3>Create an Azure Key Vault</h3><p>We’ll use Azure Key Vault to store credentials.</p><p>In this example, we’ll only use the key vault for storing our Azure Storage Account credentials. However, the key vault could be used for storing the credentials for any number of Azure resources.</p><p>Using the Azure CLI tool run:</p><pre>az keyvault create --name &lt;keyvault-name&gt; <strong>\</strong><br>    --resource-group &lt;resource-group&gt; <strong>\</strong><br>    --location &lt;location&gt;</pre><h3>Store Secrets in Azure Key Vault</h3><p>Now we can store the Azure Storage Account Key in the Key Vault. Using the Azure CLI tool run:</p><pre>az keyvault secret set --vault-name &lt;keyvault-name&gt; --name &quot;StorageAccountKey&quot; --value &lt;storage-account-key&gt;</pre><pre>az keyvault secret set --vault-name &lt;keyvault-name&gt; --name &quot;SpnPassword&quot; --value &lt;service-principle-password&gt;</pre><h3>Give service principal access to Key Vault:</h3><p>Now that we’ve got our service principal set up, we need to give this service principal access to our Key Vault</p><pre>az keyvault set-policy -n &lt;keyvault-name&gt; <strong>\</strong><br>    --spn &lt;service-principle-app-id&gt; <strong>\</strong><br>    --secret-permissions get list set delete <strong>\</strong><br>    --key-permissions create decrypt delete encrypt get list unwrapKey wrapKey</pre><h3>Give Azure DevOps Access to Key Vault</h3><p>Now we need to give Azure DevOps access to KeyVault so that our pipelines have access to Key Vault secret variables.</p><p>On Azure DevOps, navigate to “Library” under Pipelines and add a new Variable Group, you should be greeted with a screen as below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*xaGA8wS9PnQWkM9d.png" /></figure><p>Give your Variable Group a name and make sure to toggle on the button that says “Link secrets from an Azure key vault as variables”.</p><p>Then select your Azure subscription, Key vault name and add the variables you’d like to include (In this case &quot;StorageAccountKey&quot; and &quot;SpnPassword&quot;).</p><p>Click “Save” to finish.</p><h3>Add Azure Environment Variables to Azure DevOps</h3><p>Using the method above, we’ll also add our other Azure Environment Variables to KeyVault.</p><p>Here you might do this multiple times with different variables for your different environments (Dev, Test, Prod etc.). We only have one environemnt so we’ll just do this assuming our environment is the production environment.</p><p>In addition to some names of resources you defined above, you’ll also need your Tenant ID and Subscription ID at this step, this can be shown on the Azure CLI as follows:</p><pre>az account show</pre><p>The Subscription ID can be found under the key &quot;id&quot; abd the Tenant ID can be found under the key &quot;tenantId&quot;.</p><p>As above, navigate to “Library” under Pipelines and add a new Variable Group. This time don’t toggle the key vault button as you did above.</p><p>Now add variables for each of StorageAccountName, SpnID, TenantID, AmlWorkspaceName, SubscriptionID, and ResourceGroup as shown below (N.B. SpnID is the App ID for the service principle):</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ZJFFbbeV_9fd-zbE.png" /></figure><p>Click “Save” to finish.</p><h4>Data Pipeline</h4><p>This pipeline will be a data pipeline to create data and upload it to Azure Blob Storage. This datastore will then be registered with Azure Machine Learning ready for using in our model training pipeline.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*C80z7nWfXVL3AYKk.png" /></figure><p>In this example, this data will be simple and static, but this could be a continuously updating dataset that would undergo a more complex ETL pipeline at this stage that we can use to re-train models.</p><p>We’ll set this up as a daily pipeline, as if our data was being updated daily.</p><h3>Pipeline definition</h3><p>Our Azure Pipeline will be set up as a yaml file.</p><p>In our git repository, this can be found in the root of the repository as data_pipeline.yml.</p><p>Let’s take a look at this yaml file and then we’ll explain what’s happening at each step:</p><pre>trigger: none</pre><pre>schedules:<br>  - cron: &quot;0 0 * * *&quot;<br>    displayName: &quot;Daily midnight data pipeline run&quot;<br>    branches:<br>      include:<br>      - master<br>    always: true</pre><pre>name: &#39;data_pipeline&#39;<br>jobs:<br>  - job: &#39;data_pipeline_job&#39;<br>    pool:<br>      vmImage: &#39;ubuntu-16.04&#39;<br>    variables:<br>      - group: KeyVault<br>      - group: ProductionEnvVars<br>    steps:<br>      - task: UsePythonVersion@0<br>        inputs:<br>          versionSpec: &#39;3.7.6&#39;<br>          architecture: &#39;x64&#39;</pre><pre>      - script: |<br>          python -m pip install --upgrade pip<br>          pip install -r requirements.txt<br>        displayName: &#39;Install requirements&#39;</pre><pre>      - script: |<br>          python src/my_custom_package/create_data.py<br>        displayName: &#39;Create and Register Data&#39;<br>        env:<br>          STORAGE_ACCT_NAME: $(StorageAccountName)<br>          STORAGE_ACCT_KEY: $(StorageAccountKey)<br>          TENANT_ID: $(TenantID)<br>          SPN_ID: $(SpnID)<br>          SPN_PASSWORD: $(SpnPassword)<br>          AML_WORKSPACE_NAME: $(AmlWorkspaceName)<br>          RESOURCE_GROUP: $(ResourceGroup)<br>          SUBSCRIPTION_ID: $(SubscriptionID)</pre><p>So if we take a dive into at what’s happening here:</p><h4>Schedules</h4><p>First, the CI trigger is turned off, so that this Pipeline isn’t run every time the code is updated.</p><p>There is a cron schedule that runs at 00:00 every day on the master branch. This is run regardless of whether there are any code changes, because there may be data changes.</p><h4>Jobs</h4><p>We have set up a pipeline with a single stage, with a single job.</p><p>The VM image being used is an Ubuntu 16.04 image.</p><p>Variables are extracted from the variable groups we set up in our resource set up in Azure DevOps.</p><h4>Steps</h4><p>Python 3.7.6 is being used here. You can define a strategy in which you use multiple python versions (and multiple operating systems) but in this case we’re just using one version of python on one OS.</p><p>In the second step, we upgrade pip and install the requirements for our pipeline using the requirements.txt file found in the root of our repository.</p><p>In the final step we run the python file at src/my_custom_package/create_data.py in our repository. In the section below, we’ll take a look at what this file is doing.</p><p>Note at this last step that we are extracting variables from our variable groups and assigning them as environment variables for this script to use.</p><h3>Create Data</h3><p>Running the file src/my_custom_package/create_data.py will:</p><ul><li>Retrieve azure resource details and keys from the environment variables</li><li>Create our training, test and validation data sets</li><li>Upload these datasets to our Azure Storage Account</li><li>Register the datasets</li></ul><p>This is seen in our main function:</p><pre><strong>def</strong> main():<br>    # Retrieve vars from env<br>    storage_acct_name = os.environ[&#39;STORAGE_ACCT_NAME&#39;]<br>    storage_acct_key = os.environ[&#39;STORAGE_ACCT_KEY&#39;]<br>    tenant_id = os.environ[&#39;TENANT_ID&#39;]<br>    spn_id = os.environ[&#39;SPN_ID&#39;]<br>    spn_password = os.environ[&#39;SPN_PASSWORD&#39;]<br>    workspace_name = os.environ[&#39;AML_WORKSPACE_NAME&#39;]<br>    resource_group = os.environ[&#39;RESOURCE_GROUP&#39;]<br>    subscription_id = os.environ[&#39;SUBSCRIPTION_ID&#39;]</pre><pre>    # Instantiate Blob Storage Interface<br>    blob_storage_interface = BlobStorageInterface(<br>        storage_acct_name, storage_acct_key<br>    )</pre><pre>    # Create and Upload data to Blob Store<br>    data_creator = CreateClassificationData()<br>    data_creator.upload_data(blob_storage_interface)</pre><pre>    # Register Blob Store to AML<br>    aml_interface = AMLInterface(<br>        tenant_id, spn_id, spn_password, subscription_id,<br>        workspace_name, resource_group<br>    )<br>    aml_interface.register_datastore(<br>        TRAINING_CONTAINER, TRAINING_DATASTORE,<br>        storage_acct_name, storage_acct_key<br>    )</pre><p>When the CreateClassificationData() class is instantiated it creates some dummy classification data for us to use to create a classification machine learning model:</p><pre><strong>class</strong> CreateClassificationData():<br>    <strong>def</strong> __init__(self):<br>        x_arr, y_arr = make_classification(<br>            n_samples=5000,<br>            n_features=10,<br>            n_classes=2,<br>            random_state=1<br>        )<br>        col_names = [&#39;A&#39;, &#39;B&#39;, &#39;C&#39;, &#39;D&#39;, &#39;E&#39;,<br>                     &#39;F&#39;, &#39;G&#39;, &#39;H&#39;, &#39;I&#39;, &#39;J&#39;]<br>        x_df = pd.DataFrame(x_arr, columns=col_names)<br>        y_df = pd.DataFrame({&#39;Target&#39;: y_arr})<br>        # Training set n=3500<br>        self.x_train = x_df.iloc[:3500]<br>        self.y_train = y_df.iloc[:3500]</pre><pre>        # Testing set n=750<br>        self.x_test = x_df.iloc[3500:4250]<br>        self.y_test = y_df.iloc[3500:4250]</pre><pre>        # Validation set n=750<br>        self.x_valid = x_df.iloc[4250:]<br>        self.y_valid = y_df.iloc[4250:]</pre><p>We create 3 sets of data — a training set, a test set and a validation set of data. When we call the upload_data method of our CreateClassificationData class, this uploads these three sets of data to the blob store:</p><pre><strong>def</strong> upload_data(self, blob_storage_interface):<br>    self.upload_training_data(blob_storage_interface)<br>    self.upload_evaluation_data(blob_storage_interface)<br>    self.upload_validation_data(blob_storage_interface)</pre><h3>Set Up Pipeline on Azure DevOps</h3><p>First you’ll need to have set up the resources as described in part 1.</p><p>Now to set up the pipeline, first navigate to Pipelines:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*u2dMWeldGcdEvmcQ.png" /></figure><p>Then click on “New Pipeline”, you’ll be greeted with the screen below. Assuming you’re using Azure Repos git repositories, select the top option:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*U_GfOeNiqxPHyVEf.png" /></figure><p>Once you’ve selected your code source, you’ll select which repository you’re using:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CfPVM6CvTc65KMEu.png" /></figure><p>Then you need to configure your pipeline. We’ll be using a yaml file so select the bottom option below (“Existing Azure Pipelines YAML File”):</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BR9pI2JFRdcU6yh0.png" /></figure><p>When you click on this, you have the option to select a yaml file. In this case we’ll be using data-pipeline.yml, so select this yaml file:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/970/0*DKeyf7O0WtcXOHIb.png" /></figure><p>You’ll then be taken to a page in which you can review this yaml file. Once you’ve confirmed it’s all okay, click on “Run”:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*_cUasjrY9iAv5UgG.png" /></figure><p>You may need to grant your pipeline access to your subscription to access the variables it will need to access.</p><p>The pipeline will then run and, if successful, you should see an output similar to the below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/627/0*5J_30WvcMxdGZsNi.png" /></figure><h3>Rename Pipeline</h3><p>As we’ll want to name our pipeline to something more descriptive such as “Data-Pipeline”, on the Azure pipelines page, click on the ellipsis and rename the pipeline:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*XR1kzu4Tymo-QNn1.png" /></figure><h3>See Scheduled Runs</h3><p>To see the runs that are scheduled for the week, click on the pipeline, you’ll then be taken to a page as shown below. Click on the ellipsis in the top-right and click “Scheduled runs”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*4vwn-O7TCjjSiHFc.png" /></figure><p>And as same steps we will create four more pipelines that are :</p><ul><li>Environment Pipeline</li><li>Model Training Pipeline</li><li>Model Deployment Pipeline</li><li>Continuous Integration (CI) Pipeline</li></ul><p><em>Then our end-to-end MLOps in Azure will be set-up with DevOps pipelines</em></p><blockquote><strong>Thank You</strong></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f6e7c006fe0e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Path for Beginner To Expert : DevOps & Cloud]]></title>
            <link>https://medium.com/analytics-vidhya/path-for-beginner-to-expert-devops-cloud-a92b521896ad?source=rss-68478c793efe------2</link>
            <guid isPermaLink="false">https://medium.com/p/a92b521896ad</guid>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[cloud]]></category>
            <category><![CDATA[linux]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[docker]]></category>
            <dc:creator><![CDATA[Aditya Soni]]></dc:creator>
            <pubDate>Sun, 07 Jun 2020 05:56:50 GMT</pubDate>
            <atom:updated>2020-06-19T16:03:33.798Z</atom:updated>
            <content:encoded><![CDATA[<h3>Path for Beginner To Expert : DevOps &amp; Cloud</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lCP10pcWxlyhI29aZqQYgQ.png" /></figure><p>If You have similar questions like<em> </em><strong><em>How </em></strong><em>to Become </em><strong><em>DevOps &amp; Cloud</em></strong><em> Engineer? </em>, What are the tools one should learn? , It’s important to learn Ansible, Nagios, etc? and the most important question I think <strong>“From where one should start with ?”.</strong></p><h4>Then you are landed at the right place.</h4><p><strong>Hey Folks</strong> , From the past few days I got a lot of queries from undergrads or techies inquiring to shape their career towards <em>DevOps &amp; Cloud</em> and hustling to know where to get started.<em> So I came up with the idea to </em><strong><em>sum-up everything</em></strong><em> in this article that can benefit more techies with similar inquiries.</em></p><blockquote><em>I am including here path from very basic to mastery with helpful resources!</em></blockquote><p><strong>So let’s begin…</strong></p><p><strong>DevOps in digest!</strong></p><p>DevOps is a Methodology to deliver software in a better approach with automated pipelines. In this Methodology, we remove gap b/w two of teams of IT, Development team, and Operations team that’s why it terms as Dev + Ops. for more you can refer <a href="https://resources.collab.net/devops-101/what-is-devops">this</a>.</p><p><strong>Why techies are preferring and changing their domain to DevOps and Cloud?</strong></p><p>Undergrads who don’t love much competitive coding, algorithms, development stuff they easily fall and adapt <strong>DevOps and Cloud </strong>and their career, <em>( it’s not always the revelation but in most of the cases this is a harsh truth and later they begin with CP, DSA, when it comes to placements for the product based companies )</em> and <strong>as professional</strong> if you have basic CS fundamentals knowledge like OS, Networking, then you can quickly adapt if you are eager about it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/623/0*VcbTvNr0ckHIHeBJ" /></figure><p>Here’s a foresight from TOI and ET by 2022 there will be 1 million jobs in Cloud and DevOps So already a better choice.</p><p><strong>From where newbie should originate with?</strong></p><p><em>#Commands over Computer Science subjects like:</em></p><ul><li><strong>Operating System:</strong> OS acquaintance is a <strong>must </strong>for a better grasp of DevOps &amp; Cloud. Topics like: Memory Management, IPC, Virtualization, DFS, etc. one can refer to these <strong>resources </strong>for starting with <a href="https://medium.com/cracking-the-data-science-interview/the-10-operating-system-concepts-software-developers-need-to-remember-480d0734d710">10-OS-concepts</a>, <a href="https://www.tutorialandexample.com/operating-system-tutorial/">OS-TSP</a>.</li><li><strong>Linux:</strong> This is must if you commencing with <strong>DevOps &amp; Cloud</strong> one should <strong>at least </strong>have Linux architecture knowledge, FS, User groups, File permissions, package management, system monitoring, scheduling, LVM management, shell scripting, and basic commands like ls, cat, top, touch, df, etc. <em>Awareness of different flavors of it like Fedora, OpenSUSE, Redhat, Ubuntu, Kali.</em>one can refer this <strong>courses </strong><a href="https://www.youtube.com/playlist?list=PLAi9X1uG6jZ2b1mUmrUcc_aEoc8tfss8e">RHEL8_youtube</a>, <a href="https://www.udemy.com/course/complete-linux-training-course-to-get-your-dream-it-job/">Udemy</a>, or <strong>official documentations.</strong></li><li><strong>Networking:</strong> Encore its must for Connectivity and <strong>troubleshooting</strong>, discover the fundamental of subnets, routing, switching, IP, Protocols, etc there is an awesome YouTube tutorial series by <a href="https://www.youtube.com/watch?v=UXMIxCYZu8o&amp;list=PLEbnTDJUr_IegfoqO4iPnPYQui46QqT0j">Ravindrababu</a>, Along with these, one can refer this course on <a href="https://www.udemy.com/course/learning-path-linux-a-to-z-linux-networking-concepts/?utm_source=adwords&amp;utm_medium=udemyads&amp;utm_campaign=DSA_Catchall_la.EN_cc.INDIA&amp;utm_content=deal4584&amp;utm_term=_._ag_82569850245_._ad_437477497176_._kw__._de_c_._dm__._pl__._ti_dsa-506881825874_._li_9061771_._pd__._&amp;matchtype=b&amp;gclid=Cj0KCQjw_ez2BRCyARIsAJfg-kuqQ0_8rIh2DvDkeZxmCtkoXw38duu7NrcuouUkQzsp1OsNW2_RKRkaAn2REALw_wcB">Udemy</a> or <a href="https://devopshero.wordpress.com/2015/11/09/linux-networking-concepts/">Networking Concepts</a>.</li><li><strong>Database Management System</strong>: Basics of DB, SQL query, Transactions etc. here’s some useful <strong>resources </strong>to start with <a href="https://www.oreilly.com/library/view/relational-theory-for/9781449365431/ch01.html">oreilly</a>, <a href="https://www.w3schools.in/sql/database-concepts/">w3schools</a>.</li><li><strong>Programming Language:</strong> You can do better without CP here but one must have command over <strong>OOPS concepts,</strong> syntax knowledge with proficiency in any language like Python, Go, C++, etc. and I’ll highly suggest learning Python. <strong>Python</strong> is one of the primary technologies used by teams practicing <strong>DevOps and cloud</strong>, it enables building web applications, data visualizations, and also improves the workflow. one can refer to these resources <a href="https://www.udemy.com/course/complete-python-bootcamp/?utm_source=adwords&amp;utm_medium=udemyads&amp;utm_campaign=Python_v.PROF_la.EN_cc.INDIA_ti.7380&amp;utm_content=deal4584&amp;utm_term=_._ag_78980845866_._ad_437477391801_._kw__._de_c_._dm__._pl__._ti_dsa-774930046209_._li_9061772_._pd__._&amp;matchtype=b&amp;gclid=Cj0KCQjw_ez2BRCyARIsAJfg-kuyDn7XSYcAqbdyk2crYHYzE1DQW1jegEiGqMcaXS1ckLEk3541_7saAi32EALw_wcB">Udemy</a>, <a href="https://github.com/stenivan/python-zero-to-hero">Github</a>, <a href="https://www.cybrary.it/blog/0p3n/python-guide-zero-hero/">cybrary</a>.</li><li><strong>SCM tools:</strong> It is necessity to have command on <strong>VCS tools</strong> like Git, Github, BitBucket here are resources to start with <a href="https://medium.com/@javinpaul/top-10-free-courses-to-learn-git-and-github-best-of-lot-967aa314ea">Medium_blog</a>, <a href="https://try.github.io/">try.github.io</a>, <a href="https://blog.udacity.com/2015/06/a-beginners-git-github-tutorial.html">Udacity</a>.</li></ul><p><em>Oh wait you thinking these you already learned then you </em><strong><em>can start from below:</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1007/1*BpL4SAu3edB7aHiqHHZwXQ.png" /></figure><p><strong>Explore DevOps tools<em>:</em></strong> So DevOps practice we can archive with the help of orders of tools but as beginner, if you only as i have included below then good to start.</p><ul><li><strong>Docker</strong>: If you have already know the <strong>basics of OS</strong> in-depth then to learning Docker will be much simple for you similar tools like Docker are Rkt, Podman, etc but Docker is most famous and used. here are resources to start with <a href="https://docs.docker.com/">docs.docker</a>, <a href="http://dockerlabs.collabnix.com/">collabnix</a>, <a href="https://www.udemy.com/course/learn-docker/">Udemy</a>.</li><li><strong>Ansible</strong>: It’s a software provisioning, configuration management, and application-deployment tool and if you have command on this then your day to day tasks of DevOps and Cloud will be more manageable. similar tools like ansible are Puppet, chef, but ansible is most famous and used here are <strong>resources </strong>to start with <a href="https://docs.ansible.com/">docs.ansible</a>, <a href="https://opensource.com/article/19/12/ansible-resources">opensource.com</a>, <a href="https://www.ansible.com/resources/videos">RedHat’s Videos</a>.</li><li><strong>Kubernetes</strong>: K8s is container-orchestration tool and will handle your ship of DevOps, the only prerequisite is if you have learned everything <strong>till now</strong> mentioned in this article then you will easily adopt it, <em>this is a most important tool for future in DevOps</em> here are <strong>resources </strong>to start with <a href="https://kubernetes.io/">kubernetes.io</a>, <a href="http://dockerlabs.collabnix.com/kubernetes/README.html">collabnix</a>, <a href="https://kodekloud.com/">kodekloud</a>, <a href="https://www.katacoda.com/">katacoda</a>.</li><li><strong>Monitoring Tools:</strong> It’s significant to have a basic <strong>understanding </strong>of any monitoring tool. You can choose anyone among Prometheus, Grafana, etc. here are <strong>resources </strong>to start with <a href="https://prometheus.io/">prometheus.io</a>, <a href="https://grafana.com/tutorials/">grafana</a>,</li><li><strong>Jenkins</strong>: This you can say as <strong><em>Heart of DevOps,</em></strong> its an open-source automation tool built in java for Continuous Integration purposes. This tool plays a very crucial role in architecture. here are <strong>resources </strong>to start with: <strong>CI/CD</strong> with Jenkins using Pipelines by <a href="https://www.udemy.com/course/learn-devops-ci-cd-with-jenkins-using-pipelines-and-docker/#instructor-1">Edward Viaene</a>, <a href="https://www.jenkins.io/doc/tutorials/">jenkins.io</a></li></ul><p><strong>Resources Tip:</strong> <a href="https://www.katacoda.com/learn">katacoda.com</a> here you can find <strong>Hands-on</strong> questions with labs for learning and practice.</p><p><strong>Why DevOps + Cloud is a good combination?</strong></p><p><em>So till here, </em>you captured the idea about DevOps, but just imagine if any Company aspire to adopt or already adopted DevOps methodology for Production they must-have resources like Compute instances, public IPs, Domain names, Databases and Network Connectivity b/w these then it will too tough for them to have this <strong>variety of infrastructure available at office</strong> and cost to hardware for the company will be higher.</p><blockquote><em>In these type of scenario, Cloud Appears in the picture and that’s the solitary reason for thriving public cloud like AWS this much.</em></blockquote><ul><li><strong>Public Cloud: </strong>Hope you got the idea about why cloud familiarity is a must with DevOps in today’s world. I personally like <strong>AWS </strong>more in comparison with other Cloud providers. But you can always discover other cloud providers as well if you know essential concepts. here are some <strong>resources </strong><a href="https://www.aws.training/">aws.training</a>, <a href="https://acloud.guru/aws-cloud-training">acloud.guru</a>, <a href="https://inthecloud.withgoogle.com/training-discount/register.html?utm_source=google&amp;utm_medium=website&amp;utm_campaign=FY20-Q2-global-demandgen-operational-wd-training_discount&amp;utm_content=skills-campaign">cloud.withgoogle</a>.</li><li><strong>Private Cloud:</strong> Not Must but as you should ken on which use case which cloud will be the best fit, here are resources <a href="https://docs.openstack.org/horizon/latest/contributor/tutorials/index.html">docs.openstack</a>, <a href="https://www.youtube.com/watch?v=Kfj5XiNdJN0">Edureka_youtube</a></li></ul><p><strong>Projects:</strong> if you have already learned above points I mentioned then you already have enough small to big <strong>projects by uniting all tools</strong>, if you don’t have any wait from here you can check and start something <a href="https://www.youtube.com/playlist?list=PLxzKY3wu0_FJdJd3IKdiM4Om1hGo2Hsdt">valaxy technologies youtube</a>, <a href="https://www.udemy.com/course/projects-in-cloud-computing/?utm_source=adwords&amp;utm_medium=udemyads&amp;utm_campaign=DSA_Catchall_la.EN_cc.INDIA&amp;utm_content=deal4584&amp;utm_term=_._ag_82569850245_._ad_437477497176_._kw__._de_c_._dm__._pl__._ti_dsa-406594358574_._li_9061772_._pd__._&amp;matchtype=b&amp;gclid=Cj0KCQjw_ez2BRCyARIsAJfg-ktbC0Sz-JmQu7A4PaHMD-qOe4RAj4IlokF3XSDZ5_KOD-wL_iB23YYaAimSEALw_wcB">Udemy-Course</a>.</p><p><strong>Note:</strong><em> it doesn’t matter how small your project is but you should know how to explain it to someone with a decent real-life use case.</em></p><p><strong>Internship:</strong> After getting sufficient practice, next, it’s time to face the real world and their real difficulties. Working with real-time data and solving real-time problems, makes a person more self-confident about their skills so strive to start with any then further you will get more idea about it, for applying internships <a href="https://www.linkedin.com/jobs/search/?keywords=devops%20intern">LinkedIn </a>is best platform and other are <a href="https://www.indeed.co.in/Devops-Intern-jobs">indeed</a>, <a href="https://www.glassdoor.co.in/Job/devops-intern-jobs-SRCH_KO0,13.htm">glassdoor</a>, <a href="https://angel.co/company/treegain-technologies/jobs/691900-devops-engineer-intern">angel.co</a>, <a href="https://www.naukri.com/internship-jobs-in-bengaluru-bangalore?expJD=true">naukri</a>.</p><p><em>I want to add here…</em></p><blockquote><strong>The Expert Was Once A Beginner</strong></blockquote><p>So don’t ever fear by this that domain needs <strong>experience</strong>, yes they need but once you started with it then there is no comeback you gonna <strong>love </strong>this.</p><p><strong><em>If you think this article can help someone then share with them, and show some love by claps..</em></strong></p><p><em>Any Suggestions or query are welcome!.. You can Connect with me at </em><a href="https://www.linkedin.com/in/theadisoni/"><em>LinkedIn</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a92b521896ad" width="1" height="1" alt=""><hr><p><a href="https://medium.com/analytics-vidhya/path-for-beginner-to-expert-devops-cloud-a92b521896ad">Path for Beginner To Expert : DevOps &amp; Cloud</a> was originally published in <a href="https://medium.com/analytics-vidhya">Analytics Vidhya</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>