<?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[alter-way-innovation - Medium]]></title>
        <description><![CDATA[French Open Source Expertise - Medium]]></description>
        <link>https://medium.com/alter-way-innovation?source=rss----65f4595d7c5d---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>alter-way-innovation - Medium</title>
            <link>https://medium.com/alter-way-innovation?source=rss----65f4595d7c5d---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 03:02:47 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/alter-way-innovation" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Javascript charts comparator]]></title>
            <link>https://medium.com/alter-way-innovation/javascript-charts-comparator-8ddaf6816ac3?source=rss----65f4595d7c5d---4</link>
            <guid isPermaLink="false">https://medium.com/p/8ddaf6816ac3</guid>
            <category><![CDATA[chartjs]]></category>
            <category><![CDATA[apexcharts]]></category>
            <category><![CDATA[chartist]]></category>
            <category><![CDATA[comparison]]></category>
            <category><![CDATA[uitoast]]></category>
            <dc:creator><![CDATA[Koku Ulrich GBLOKPO]]></dc:creator>
            <pubDate>Thu, 02 Dec 2021 14:49:39 GMT</pubDate>
            <atom:updated>2021-12-02T15:04:30.226Z</atom:updated>
            <content:encoded><![CDATA[<p>As nowadays applications resolve more and more on displaying numeric data, the need for charts libraries is growing accordingly. In the Javascript ecosystem, there are few libraries which each have their own subtleties, sometimes style wise, sometimes features wise.</p><p><strong>Javascript charts comparator </strong>is a Storybook implementation, that presents the most used libraries in the field, so developers can see the representation of the same chart type over various libraries and thus be able to choose the right library for their functional need.</p><p><strong>Interact and compare</strong></p><p>We focused on four libraries which are among the most used open source charts libraries in javascript : <a href="https://apexcharts.com/">Apexcharts</a>, <a href="https://www.chartjs.org/">Chart.js</a>, <a href="https://ui.toast.com/">Ui.Toast</a> and <a href="https://gionkunz.github.io/chartist-js/">Chartist</a>.</p><p>We propose two level of information, first through an interactive Storybook that holds chart examples under few categories : Area charts, Bar charts, Box charts, Donut &amp; Pie…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/896/1*xPJTeZ-FMCeD1x5j1vUsfg.gif" /></figure><p>Second, we built a small comparison table for the libraries through their features.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/878/1*RbaqzCBduYv--WnTxuMZUw.png" /><figcaption>Open source javascript chart libraries offer customization, animations and responsiveness out of the box.</figcaption></figure><h4>Get the code</h4><p>Project is available over Github : <a href="https://github.com/alterway/javascript-charts-comparator">https://github.com/alterway/javascript-charts-comparator</a></p><p>Follow the readme to launch the app and make your comparison !</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8ddaf6816ac3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/alter-way-innovation/javascript-charts-comparator-8ddaf6816ac3">Javascript charts comparator</a> was originally published in <a href="https://medium.com/alter-way-innovation">alter-way-innovation</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Transform your system metrics into KPIs with the  Rating Operator !]]></title>
            <link>https://medium.com/alter-way-innovation/transform-your-system-metrics-into-kpis-with-the-rating-operator-588bb994520d?source=rss----65f4595d7c5d---4</link>
            <guid isPermaLink="false">https://medium.com/p/588bb994520d</guid>
            <category><![CDATA[iaas]]></category>
            <category><![CDATA[metrics]]></category>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[cloud-cost]]></category>
            <category><![CDATA[billing]]></category>
            <dc:creator><![CDATA[Sihem Cherrared]]></dc:creator>
            <pubDate>Wed, 27 Oct 2021 13:11:08 GMT</pubDate>
            <atom:updated>2021-11-02T16:15:56.061Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qjw9-9gt8-V9LUENIQPjdw.png" /></figure><h3>Transform your system metrics into KPIs with the Rating Operator !</h3><p>Rating Operator is a Kubernetes based application aimed at providing system administrators, DevOps or business-oriented users, such as BI developers, with indicators or KPIs (Key Performance Indicators) derived from metrics.</p><p>The need for KPIs can be manyfold, from saving storage space by aggregating metrics over time and environments, to providing higher level or domain oriented information over the ongoing operations. These information can be cost, energetic use, performance analysis, at real-time or forecast…</p><p>Whereas the classical scheme for a DevOps engineer is to build its own KPIs by defining a PromQL (Prometheus), LogQL (Loki) or InfluxQL (InfluxDB) queries and then visualizing them in Grafana, Rating Operator API helps into defining transformation rules, that will target metrics, define the mathematical operations to be performed on them, so to output a domain-oriented KPI. Moreover, Rating Operator enables users that are non familiar with PromQL to define their own KPIs using existing transformation rules templates or to create new ones. These templates can be defined by developers and applied by clients of any business domain.</p><p>In order to define KPIs, the Rating Operator API applies three rating rule objects: templates, values and instances. The rating rule template is a PromQL rule defined by the developer. This template contains a list of variables that are set to specific values to create the desired KPI (i.e. rating rule instance).</p><p>Suppose, you want to calculate the cost of your kubernetes nodes if you deploy your infrastructure in AWS and apply the <strong>aws_a1_large flavor</strong>.</p><p>To do so you need to create a PromQL rule that considers the cpu and memory consumption of Kubernetes nodes and apply the price of the <strong>aws_a1_large </strong>flavor. To make this PromQL reusable for other flavors, you can create a rating rule template using the Rating Operator API:</p><pre>- <strong>query-name</strong> : kubernetes-nodes-cost<br>-<strong> query-group</strong> : cloud-cost <br>- <strong>query-template</strong> :(ceil(sum(instance:node_memory_utilisation:ratio)   * max(node_memory_MemTotal_bytes)/(1024*1024*1024)))/sum({nb_ram}) &gt; (ceil(sum(instance:node_cpu:ratio) * ax(instance:node_num_cpu:sum))) / sum({nb_cpu}) or (ceil(sum(instance:node_cpu:ratio) * max(instance:node_num_cpu:sum)))/sum({nb_cpu})) * sum({price})</pre><p>The created <em>kubernetes-nodes-cost</em> template can then be applied for any flavor in AWS. If we consider the example of <strong>aws_a1_large</strong> flavor, we can then specify its size and price values as a rating rule valuesobject. The <strong>aws_a1_large </strong>values are defined as metrics and the modification history is stored in the database for further use. The Rating Operator API creates the rating rule instance according to the values you specified fo the<strong> aws_a1_large </strong>flavor:</p><pre>Name: aws-a1-large-flavor<br>Namespace: rating API <br>Version: rating.alterway.fr/v1 <br>Kind: RatingRule <br>Spec: <br>Metric: ((ceil(sum(instance:node_memory_utilisation:ratio) * max(node_memory_MemTotal_bytes)/(1024*1024*1024)))/sum(2) &gt; (ceil(sum(instance:node_cpu:ratio) * max(instance:node_num_cpu:sum))) / sum(4) or (ceil(sum(instance:node_cpu:ratio) * max(instance:node_num_cpu:sum)))/sum(4)) * sum(0.051) <br>Timeframe: 3600</pre><p>You can then visualize your results in Grafana or apply other flavors with the same template.</p><h3>Grouping templates and metrics:</h3><p>In the Rating operator, the templates and metrics are organized into groups to simplify their reusability. The following figure showcases the rating rule template hierarchy with the example cited above.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/873/1*4fe6028ZPY56S0x0Zs-NcQ.jpeg" /><figcaption>Rating operator templates and metrics grouping</figcaption></figure><h3>Rating Operator architecture:</h3><p>The Rating Operator architecture is divided into three main components that interact with Prometheus, Grafana and Kubernetes APIs:</p><ul><li>Rating Operator API: proposes a list of endpoints to create and modify your KPIs.</li><li>Rating Operator Manager: watches and updates CustomResources, helping other components achieve their roles. It plays a role in most of the features of the rating-operator, handling all event based callbacks.</li><li>Rating Operator Engine: uses custom resources to configure its workers and trigger rating workloads, with a variable timeframe. Each worker holds its own configuration, that can be updated at runtime, and query, rate and store data in PostgreSQL through the Rating operator API.</li></ul><p>Natively, all the data generated by rating mechanisms is accessible by the system administrator. If the context requires data segregation, we provide a mechanism (Local, Keycloak or using LDAP schema), to authenticate your users (tenants) and assign them namespaces. A logged in tenant can only access data coming from its namespaces. Any tenant registered to the application has a Grafana account with the same credentials created. In Rating Operator, we provide three ways to authenticate users:</p><ol><li><strong>Local authentication</strong> using the PostgreSQL database.</li><li><strong>Keycloack</strong>: an open source identity and access management.</li><li><strong>Lightweight Directory Access Protocol (LDAP)</strong>: compatible with LDAP open source solutions such as OpenLDAP.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UO-0TPaKcZsUNngkInGoNA.jpeg" /><figcaption>Rating operator architecture</figcaption></figure><h3>Rating Operator use cases and upcoming features:</h3><p>Rating Operator proposes a set of ready to use dashboards and templates. For instance, the cloud costs comparison dashboard that allows you to compare the costs of your kubernetes infrastructure resources consumption over different IAAS cloud providers. One of the last added features is the automatic rules detection, that enables you to find templates that match your needs with machine learning.</p><p>Follow us on Github, we need your feedbacks: <a href="https://github.com/alterway/rating-operator">https://github.com/alterway/rating-operator</a></p><p>Rating Operator was made in Alter Way with ❤ by its R&amp;D team and contributors (<a href="https://medium.com/u/f9b50ae3c36a">Jonathan Rivalan</a>, <a href="https://medium.com/u/324b115d428a">Sihem Cherrared</a>, <a href="https://medium.com/u/f045bac7eb29">Phuc Vu</a>, <a href="https://medium.com/u/ad7506070d02">Valentin Daviot</a>)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=588bb994520d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/alter-way-innovation/transform-your-system-metrics-into-kpis-with-the-rating-operator-588bb994520d">Transform your system metrics into KPIs with the  Rating Operator !</a> was originally published in <a href="https://medium.com/alter-way-innovation">alter-way-innovation</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Automatic qualification plugin for Redmine]]></title>
            <link>https://medium.com/alter-way-innovation/automatic-qualification-plugin-for-redmine-a57a6f8c20d1?source=rss----65f4595d7c5d---4</link>
            <guid isPermaLink="false">https://medium.com/p/a57a6f8c20d1</guid>
            <category><![CDATA[redmine]]></category>
            <category><![CDATA[workflow-automation]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Veeraraghavan SEKAR]]></dc:creator>
            <pubDate>Tue, 17 Aug 2021 12:19:01 GMT</pubDate>
            <atom:updated>2021-08-20T14:31:39.829Z</atom:updated>
            <content:encoded><![CDATA[<h3>Redmine and pipelines</h3><p><a href="https://www.redmine.org/"><em>Redmine</em></a> is a fantastic open source tool allowing teams and companies to create, list, prioritize and manage issues. If you know about <a href="https://www.atlassian.com/software/jira"><em>Jira</em></a>, <em>Redmine</em> is its open source counterpart. While being extremely minimalist it is also highly modular. So when we had to extend our internal support system with automatic issues processing, we directly thought of our solution as a <em>Redmine </em>extension.</p><p>We looked at a way to create a simple and flexible interface to define arbitrary workflows when issues are created. Moreover, in order not to execute any arbitrary code over our <em>Redmine</em> instance, we imagined a web hook system. Then we decided to chain those calls to be able to use third-party web services together.</p><p>For the oldest readers, we had <a href="https://en.wikipedia.org/wiki/Yahoo!_Pipes"><em>Yahoo Pipes</em></a> in our mind. <em>Yahoo pipes</em> product was an innovative solution from the early 2000’s, bringing syndication systems such as <a href="https://www.netvibes.com/en"><em>Netvibes</em></a> to another level. <em>Yahoo Pipes</em> used to propose a web based graphical user interface that enabled its users to defines pipes made of various RSS services, custom scripts... Glued together, those services and scripts could provide any kind of information filtering pipeline. The service was made free, ans as it was not sustainable it fatally closed in 2015.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QY8EgzlV6LTxuFcTMVm9eA.png" /><figcaption>Yahoo Pipes example</figcaption></figure><p>The concept stayed vivid in the minds of many leading to the creation of paid services like <a href="https://ifttt.com/"><em>IFTTT</em></a>, <a href="https://zapier.com/"><em>Zapier</em></a> or self hosted alternatives like <a href="https://github.com/huginn/huginn"><em>Huginn</em></a>. Since we had to bind our solutions with <em>Redmine</em> internal hooks, we developed our own pipeline system has a <a href="https://medium.com/r?url=https%3A%2F%2Fgitlab.com%2Falter-way-rnd%2Fredmine-qualification-plugin"><em>Redmine</em> plugin</a>. Our solutions calls iteratively one to many services upon issue creation, after every step we save the textual response of the service in a Redmine custom field to be eventually reused later.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ScNyC0BJlUpW3v6UJgrV-w.png" /><figcaption>Configuring Individual Service in Plugin</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/841/0*wGEVrei1pka56UFm.PNG" /><figcaption>Plugin configuration for each project</figcaption></figure><p>We added the possibility to enable or disable the plugin entirely or some of its pipelines for each project. This allows users to customize the behavior of their workflow directly from the project configuration. In the end, as many projects can have as many pipeline configurations, depending on their business and teams needs or constraints.</p><h3>Use case and new plugin version</h3><p>What are we trying to achieve at Alter Way ? Behind this plugin lies a machine learning project, we described more extensively in <a href="https://medium.com/alter-way-innovation/nath-for-redmine-nlp-based-tickets-classification-a98cac2bb304">a previous article</a>. To put it short, we aim to automatically qualify incoming issues by asking our algorithms the following questions :</p><ul><li>Is it a fix or a new feature ?</li><li>Is this issue critical ?</li><li>How long will it take to be solved ?</li><li>Do we have similar issues ?</li></ul><p>This analysis, also known as <em>triage</em>, is usually performed by our support team but it is time-consuming. Moreover we can hardly rely on human agents (be them client side or support one) to precisely answer them, as they are experienced based, especially for time estimation and similarity.</p><h3><strong>Introducing Qualification plugin</strong></h3><p>To achieve this use case, we published three years ago an initial <em>Redmine</em> plugin called <a href="https://gitlab.com/alter-way-rnd/NATH/NATH-Redmine">NATH Redmine</a>. This plugin used web hooks but in a constraining way. It required the service to accept strict answers and responses format, which did not allow for services chaining. The machine learning ecosystem evolving so quickly, those constraints ended up being hard limitations. We needed a flexible solution capable of following our prototypes, and this his why we created <a href="https://gitlab.com/alter-way-rnd/redmine-qualification-plugin"><strong><em>Redmine-qualification-plugin</em></strong></a><strong><em> </em></strong>, a flexible way to automate your Redmine workflows.</p><p>The plugin is licensed under MIT and is open to pull requests.</p><p>Thanks to Alter Way, and the RnD team for making this project possible.</p><p>Keep innovating!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a57a6f8c20d1" width="1" height="1" alt=""><hr><p><a href="https://medium.com/alter-way-innovation/automatic-qualification-plugin-for-redmine-a57a6f8c20d1">Automatic qualification plugin for Redmine</a> was originally published in <a href="https://medium.com/alter-way-innovation">alter-way-innovation</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Teach your AI to say NO]]></title>
            <link>https://medium.com/alter-way-innovation/teach-your-ai-to-say-no-454e9146d4d8?source=rss----65f4595d7c5d---4</link>
            <guid isPermaLink="false">https://medium.com/p/454e9146d4d8</guid>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Damien GILLES]]></dc:creator>
            <pubDate>Thu, 29 Aug 2019 14:21:17 GMT</pubDate>
            <atom:updated>2023-04-03T11:23:41.760Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*dS-871ArH_afIW3L.png" /></figure><p>Automation has taken an ever growing place in the society, but we are facing a new problem: can we trust those automated systems? Reducing to the bare minimum the error rate and detecting edge cases become more and more important as our AIs takes part in more critical missions. Moreover, humans tend to act blindly in front of the machines’ decisions, making them potentially even more catastrophic. To mitigate those issues we researched a way for the AIs to call for human help.</p><h4>Our use case</h4><p>Alter Way offers application maintenance services to their client. The current workflow requires the clients to fill a form by filling a category (demand or anomaly), a title and a description. Then, the issue is further classified and resolved by our agents as discussed in our <a href="https://medium.com/alter-way-innovation/nath-for-redmine-nlp-based-tickets-classification-a98cac2bb304">previous article</a>.</p><p>To improve our workflow we work on automating most of the classifications. However, we must ensure that the tickets are correctly classified to prioritize the issues and deal with the critical ones in priority.</p><h4>A third class: “human”</h4><p>We reali<em>z</em>ed early on that automating our process may lead to mistakes and that the operators are unlikely to edit the classifications once a prediction as been made. In this context, we explored ways to call for human help when our algorithm was uncertain.</p><p>Our algorithm’s binary classifier is based on a stochastic gradient descent, so it returns the probability of belonging to a class. If the probability is over 50% the ticket is tagged with that class. The naive approach to discriminate cases where our algorithm isn’t sure how to classify an issue is to create a third class for the tickets which are close to 50% chance to belonging to a class. However, this approach throughput many errors on issues that our algorithm just does not understand.</p><p>We found a way out while working with <a href="https://openreq.eu/">OpenReq</a>, by exploring and benchmarking multiple algorithms that you can find over <a href="https://github.com/OpenReqEU">their github</a>. We realized that while both our and their algorithms have about the same precision, they made quite different mistakes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/727/1*j2GzaZR7pKMirN2-jl3V1w.png" /></figure><p>By combining the naive approach with the opposition between both algorithm results we improved our error rate dramatically. <strong>Indeed, our error rate was reduced by 63% by adopting this workflow</strong> (SS is Alter Way, RI is OpenReq):</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/891/0*K1x0zeXd13FncrGF.png" /></figure><p>This decision tree reduces our error rate to 3%, while requiring human assistance only 5% of the time.</p><h4>In a nutshell</h4><p>We managed to gain trust in our system by allowing it to refuse to qualify an issue. When we do so, we require human qualification to fully qualify the issue. This mechanism reduced our error rate by 5% in comparison to our initial algorithm. This kind of cross-validation is a security mechanism we must implement to avoid inexplicable and potentially harmful decision by our automated systems.</p><p>Thanks to Jonathan Rivalan for making this topic possible.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=454e9146d4d8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/alter-way-innovation/teach-your-ai-to-say-no-454e9146d4d8">Teach your AI to say NO</a> was originally published in <a href="https://medium.com/alter-way-innovation">alter-way-innovation</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[NATH for Redmine: NLP-based tickets classification]]></title>
            <link>https://medium.com/alter-way-innovation/nath-for-redmine-nlp-based-tickets-classification-a98cac2bb304?source=rss----65f4595d7c5d---4</link>
            <guid isPermaLink="false">https://medium.com/p/a98cac2bb304</guid>
            <category><![CDATA[issue-tracker]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Damien GILLES]]></dc:creator>
            <pubDate>Mon, 16 Jul 2018 14:02:05 GMT</pubDate>
            <atom:updated>2018-07-26T12:05:55.282Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZSSzj7CBndILIYRXY42RPw.png" /></figure><h3>Automatic issues analyzer</h3><p>Hi everyone, I am Damien Gilles, engineer for the R&amp;D team at alter way and developer of the NATH (NLP Automated Ticket Handler) for Redmine. In this article, I will explain to you how we ended up using artificial intelligence to better qualify our customers issues through <a href="https://gitlab.com/alter-way-rnd/NATH/NATH-redmine"><strong>NATH Redmine</strong></a> implementation, and why we went from there to an experimental Gitlab implementation.</p><h3>No business pain, no gain</h3><p>When first discussing with the support team, we had a strong belief that a conversational UI along with a chatbot would make their life easier, by separating the customers issues expression with the actual tracking and resolving of the issue.</p><p>In fact, what appeared to be a painful task was to qualify the customer demand within the triage phase, first by distinguishing what was an anomaly or an evolution, then by determining its criticality and where it should stand in the support team backlog. So, we considered focusing on the current Redmine workflow rather than providing a yet to be developed interface.</p><h3>Redmine and custom fields</h3><p>Some examples of fields we fill for each ticket may include:</p><ul><li>tracker: the tracker in which the ticket will appear</li><li>status: is the ticket open, close, ongoing…</li><li>urgency: how fast a ticket must be solved</li><li>importance: how big are the implications</li><li>estimated time: how long will it take to solve</li></ul><p>This classification helps us stay on track, but it also comes with multiple drawbacks:</p><ul><li>It’s time-consuming</li><li>Clients may abuse fields such as “urgency”</li><li>Agents don’t like to specify the estimated time</li><li>It’s repetitive</li></ul><h3>Classification using AI</h3><p>Thankfully, all those drawbacks can be overcome using AI. Indeed, all those fields can be deduced accurately using natural language understanding (NLU) or regression models, equivalent to the ones used for sentiment analysis. This approach allows us to save human resources and treat the tickets faster by automatically classifying them at their creation.</p><p>With modern NLU services like LUIS, Dialogflow or wit.ai you can create models for fields such as “importance” without any knowledge of machine learning. Our plugin’s flexible endpoint configuration allows us, to easily bind those service to Redmine. If you wish to explore this path we offer a <a href="https://gitlab.com/alter-way-rnd/NATH/NATH-redmine"><strong>quick walk-through</strong></a> to bind Redmine and LUIS.</p><p>However, it may be harder to find existing service to determine continuous fields, like the “estimated time”. For this one we chose a more classical approach: we built a linear classifier and published it on a local server. If you are interested in how to build such a classifier, <a href="https://www.twilio.com/blog/2017/12/sentiment-analysis-scikit-learn.html">this tutorial</a> is a good starting point. We also provide a way to run our own model <a href="https://gitlab.com/alter-way-rnd/NATH/NATH-redmine/tree/master/extra">here</a>. This simple approach can give surprisingly accurate results as illustrated by the following graph, where we compared human and machine results on 82 tickets:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KwObJurtE7ePs8R3VOHcgg.png" /></figure><p>After reviewing the few anomalies, we’ve concluded that they were mostly human errors and that the model’s prediction may be preferable to the human ones.</p><h3>Redmine integration</h3><p>We have managed to easily create some powerful IA but we still have to integrate them in our workflow and here comes NATH Redmine! Redmine does not offer an interface to configure webhooks but only provides hooks for plugins. So we’ve decided to build our own. The plugin allows you to call multiple webhooks on a ticket’s creation and to modify the value of a Redmine native or custom field.</p><p>Concretely, if you wish to automatize the custom field “estimated time” like in the example above, all you have to do is:</p><ol><li>Go to the plugin configuration page</li><li>Provide the endpoint address for your service in the text box for “estimated time”</li><li>Specify where the information is in the JSON response in “response path”</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*lxPt4lKu9PfcW9XuDqQfHw.gif" /></figure><p>This way you can bind your Redmine issue field with any web service with an HTTP GET endpoint returning JSON within minutes.</p><h3>Toward a Gitlab implementation</h3><p>As you may well know, Gitlab offers a rather complete tracking features, including binding issues status to commits labels. Following this trend, we considered implementing our algorithms within it, also in the idea of opening the platform to customers rather than only to developers.</p><p>While providing an easy to use webhooks system to trigger actions on the platform, the issue feature is lacking the possibility to implement custom fields… at all. Rather than implementing a beta feature out of our scope, we used labels to provide about the same level of functionality as our Redmine plugin does.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1yqML42Tr8_WahwYB75mkQ.png" /></figure><h3>Follow up</h3><p>We will Open Source the Gitlab plugin in the coming weeks, before proceeding with more advanced issues automation, such as automated pointers to existing solutions and in the best of our hopes, a first step toward automated resolution.</p><h3>Use and Contribute</h3><p>Meanwhile, if like us you wish to increase your reactivity when receiving tickets and likes to be ahead of time, feel free to add the <a href="https://gitlab.com/alter-way-rnd/NATH/NATH-redmine"><strong>plugin</strong></a> to your Redmine. The <a href="https://gitlab.com/alter-way-rnd/NATH/NATH-redmine"><strong>plugin</strong></a> is licensed under MIT and can be activated one project at a time, without disturbing your current workflow.</p><p>There is still room for improvement, so if you have any proposals be sure to share them with us on <a href="https://gitlab.com/alter-way-rnd/NATH/NATH-redmine/issues"><strong>gitlab</strong></a>.</p><p>Thanks to Jonathan Rivalan for making this project possible.</p><p>Keep innovating!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a98cac2bb304" width="1" height="1" alt=""><hr><p><a href="https://medium.com/alter-way-innovation/nath-for-redmine-nlp-based-tickets-classification-a98cac2bb304">NATH for Redmine: NLP-based tickets classification</a> was originally published in <a href="https://medium.com/alter-way-innovation">alter-way-innovation</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Drill down your infrastructure with Benchdrill]]></title>
            <link>https://medium.com/alter-way-innovation/drill-down-your-infrastructure-with-benchdrill-a9906187e8d7?source=rss----65f4595d7c5d---4</link>
            <guid isPermaLink="false">https://medium.com/p/a9906187e8d7</guid>
            <category><![CDATA[docker]]></category>
            <category><![CDATA[containers]]></category>
            <category><![CDATA[alter-way]]></category>
            <category><![CDATA[metrology]]></category>
            <category><![CDATA[benchmark]]></category>
            <dc:creator><![CDATA[Tasty Tomato]]></dc:creator>
            <pubDate>Mon, 28 Aug 2017 12:07:48 GMT</pubDate>
            <atom:updated>2017-08-28T12:13:27.609Z</atom:updated>
            <cc:license>https://creativecommons.org/licenses/by-sa/4.0/</cc:license>
            <content:encoded><![CDATA[<p><a href="https://github.com/Wolphin-project/Benchdrill">A benchmark-tools tool for distributed systems</a></p><p>Since we can’t automatically generate bug-free software, we have to test the software we write. Also we have to keep an eye on our applications’ performance. Benchmarks are a handy way to help you achieve this goal. Although there are several amazing tools in the outside that fill this need, we met a use case that called for a new application: <a href="https://github.com/Wolphin-project/Benchdrill">Benchdrill</a>.</p><p>The main reason to build this tool was to be able to easily charge the hosts in a Docker Swarm. Indeed, Alter Way is developing a <a href="https://github.com/Wolphin-project">big and complex project</a> which involves a metrology stack. The issue for us was a lack in the ability to test the correctness of this stack, as well as having a way to examine its charge on the metrics its collecting. To achieve these goals, we needed a software with these features:</p><ul><li>be usable with a simple CLI (no GUI);</li><li>charge CPU, I/O disks, and network of several hosts in the Docker Swarm, and all of that in parallel;</li><li>have these jobs done quickly, without too much overheard (which forced us to NOT create a container per job, as container creation weight on the system).</li></ul><p>Rapidly, we decided to use <a href="https://github.com/RichardKnop/machinery">Machinery</a> since it’s architecture is well-suited to our use case. It allows to sends tasks through a queue to a message broker, which will then distribute them to the workers. When the workers finish their jobs, they send their outputs to a result backend which will display them in the terminal afterwards. A number of message brokers are available, from which we choose <a href="https://redis.io/">Redis</a> since it can also be used as the result backend, which make things simpler.</p><p>So we have this nice tool, how to use it with Docker now? Simply by containerising the workers (no need to do that for Redis since it has an official image on the Docker Hub). After writing the Dockerfile, the following question raised: how to distribute the workers in the Swarm? By using a Compose file defining a service for workers, and another for Redis. This way, the number of workers are determined by the scale option of the docker service command. Want 100 workers? Typedocker service scale benchdrill_worker=10,easy peasy!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*72MbqlrDP94SwYLTuubbAg.gif" /></figure><p>Now we have workers distributed in the Swarm, ready to receive their tasks and execute them diligently. Now what tasks will we send them actually? Charging loads on a machine ressources is not an easy task, but thankfully there are already awesome benchmark tools that fit this complex work. The two we already integrated in the workers are Sysbench and Filebench. <a href="https://github.com/akopytov/sysbench">Sysbench</a> comes with multiple built-in benchmarks, but its strong point is that it scriptable with Lua. Although, it is recognized as a good tool to benchmark databases, this comes with the downside that you don’t have a complete control of what’s going on in the system you are testing. For example, if you want to test the I/O disk, you have to consider all the caching that is involved in modern systems. For this specific task, <a href="https://github.com/filebench/filebench/wiki">Filebench</a> is the masterpiece benchmark tool to use. It has indeed a lot of options to define what you precisely want to do when testing I/O. Of course, this comes with a higher complexity, as it can be seen with the WML, the language that is used to described the tests you want to run.</p><p>To integrate these two benchmarks, the idea was simple: write a task that encompasses the command used normally with Sysbench or Filebench, and transmit it to the workers. Although it worked well with Sysbench, that wasn’t the case with Filebench. Indeed, since the last year, the developers decided to drop the ability to write a test directly in the terminal. They explained that it was too error-prone for new users but also for experimented ones, and it’s not really ergonomic either. So you can only pass a file to the CLI of Filebench, which will read it and execute it. The workaround for that issue is to pass a file to Benchdrill CLI, which will read it and transmit its content as a string object to the workers. They will write it in a local file again then pass it to Filebench as an argument. This is realized by a different task than the previous. Of course, you can also use this task for Sysbench if you want to pass a file.</p><p>The next step was to deploy Benchdrill services in a Swarm that contains the metrology stack, then look at the results of the loads charged with Benchdrill in a Grafana dashboard. We used an infrastructure composed of 4 computers. They all have 32 GB and an 8 cores CPU. One of this CPU has a clock rate of 2.66 GHz while the 3 others has a clock rate of 3.00 GHz. Their storage capacity is 512 GB.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ln3BxdFUTaCgtG-8lkbJ9g.png" /></figure><p>This dashboard shows the total CPU time consumed by a Docker service, information retrieved with InfluxDB. In this case, we see, with purple curve, that the charge triggered by 20 workers executing 100 Sysbench CPU built-in test in parallel . Other curves depict Redis and metrology’s consumption of charge. We observe that despite the increase of the CPU time used by the workers, other services don’t show any modifications in their behaviour. Which is exactly what we want, because it means that the metrology stack charge doesn’t depend of the activity of the services it monitors.</p><p>Now, the next step of the big project is the development of the orchestration stack. Benchdrill will be helpful, because it can be used to charge heavily some nodes of the cluster while others would not. Therefore, we can observe how the orchestration stack will handle this situation, e. g. if it will balance the overall load of the cluster.</p><p>Found it intriguing? Benchdrill is a free software, so don’t hesitate to contribute in the <a href="https://github.com/Wolphin-project/Benchdrill">GitHub page</a> or open an issue to ask a question or report a bug! We’ll happily collaborate with you!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a9906187e8d7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/alter-way-innovation/drill-down-your-infrastructure-with-benchdrill-a9906187e8d7">Drill down your infrastructure with Benchdrill</a> was originally published in <a href="https://medium.com/alter-way-innovation">alter-way-innovation</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Announcing : AiC 2.0, an Open Source Android cloud emulator]]></title>
            <link>https://medium.com/alter-way-innovation/announcing-aic-2-0-an-open-source-android-cloud-emulator-8571b06c424d?source=rss----65f4595d7c5d---4</link>
            <guid isPermaLink="false">https://medium.com/p/8571b06c424d</guid>
            <category><![CDATA[emulation]]></category>
            <category><![CDATA[android]]></category>
            <category><![CDATA[openstack]]></category>
            <category><![CDATA[virtualization]]></category>
            <category><![CDATA[alter-way]]></category>
            <dc:creator><![CDATA[Jonathan Rivalan]]></dc:creator>
            <pubDate>Fri, 19 May 2017 14:37:52 GMT</pubDate>
            <atom:updated>2017-05-22T15:33:42.884Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/957/1*7lYRAmqR_GRJktGEHe2hmg.png" /></figure><p>Hello there, I am Jonathan Rivalan, project manager for the Open Source Android cloud emulator : AiC 2.0. And today, <a href="https://aic-project.github.io/">along with the sources being published</a>, i’m gonna drive you quickly through its main features and the background story behind the project.</p><h3><strong>Android versions x Compatible devices x Tests to be run = Fragmentation issue</strong></h3><p>Every single developer can understand how hard it is to develop applications expected to work seamlessly on numerous versions of an OS, deployed on numerous devices. Especially when you need to run tests for each potential combinations.</p><p>Google makes a dashboard available as part of the Android developer documentation showcasing <a href="https://developer.android.com/about/dashboards/index.html">current use of Android versions</a>. At the time of this article, 11 minor versions are in use, spreaded over 5 main versions (ranging from 2.x to 7.x).</p><p>Coming to physical devices, Google published a document listing all of <a href="https://support.google.com/googleplay/answer/1727131?hl=en">Google play supported devices</a>, 242 pages long, and summing up for a whooping 14K+ supported devices.</p><p>Even assuming a mere 10% is in use, that would be more than 1000 devices to maintain, which would need a complete reinstall if you‘re in need to rerun your tests campaign on a different Android version. Making it hard if not impossible for editors to drive in-house large scale tests campaigns.</p><h3>The Saas version of Android</h3><p>In 2012 when first specifying AiC 2.0, mostly 3 solutions were available for emulating Android. The <a href="https://developer.android.com/studio/index.html">Android Studio</a> suite developed and maintained by Google and the community, <a href="https://www.genymotion.com/">Genymotion</a> which is a successful, developer acclaimed product published with closed sources and <a href="http://www.android-x86.org/">Androidx86</a>, technically not emulating but porting the Android system to x86, when it usually is running on ARM architecture.</p><p>All these solutions had at least 2 things in common, they were local, provided a single Android instance at a given time, and had a strong tie with the user system hardware in terms of performance. Not every box could at this time emulate the last version of Android KitKat on let’s say a brand new Nexus 4 config, not to mention support for graphical hardware acceleration.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*aXA9ttaqTIPrwsEdQ6dlzg.gif" /><figcaption>Android right into your web browser.</figcaption></figure><p>So we<strong> </strong>roll up your sleeves<strong> </strong>and one step at a time aimed to send Android in the cloud. Offering what virtualization had best to offer : physical layer abstraction for both the Android devices and the user hardware, delivering Android instances right into developers browsers.</p><h3>Continuous Integration</h3><p>Although our main aim was to provide an easy, hardware dependency free, access to Android instances, we did consider to implement mechanisms project-management oriented.</p><p><strong>Tests description</strong><br>We provide an easy to use test expression system, built on top of UIAutomator. Following the Gherkin philosophy, it enables writing user stories in a natural way, which can both set the sensors at desired values and perform functional actions and controls.</p><p><strong>Asynchronous campaign</strong><br>Finally, our users can configure and launch asynchronous campaigns, that run tests (AiC2.0 natives or UIAutomator ones) on a specified Android versions set. Each Android instance is accessible while the campaign is running, and once ran, user is granted access to a detailed stack trace, enlighting what went wrong.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*9nFUOOE1BtS-5XexZa21QA.gif" /><figcaption>2 different Android releases in 2 different flavors (tablet &amp; phone), tested at the same time.</figcaption></figure><h3>A-Team effort</h3><p>AiC2.0 is not a one company effort, on the opposite it united industrials (alter way, zenika, thalès) to work along with academics (lip6, liafa) for 3 years. To provide a bit more background, AiC2.0 was a french project financed thanks to a public innovation program known as the FUI and sponsored by Systematic GTLL cluster.</p><h3>Use and Contribute</h3><p>If you read this article up to this point then you may have some interest in either using the solution or contributing to it. We are open to both proposals. In cooperation with OneLab we are now opening access to test the platform hosted at UPMC (Université Pierre et Marie Currie). Create a OneLab account (and select “alterway” as an organization) <a href="https://portal.onelab.eu/register">https://portal.onelab.eu/register</a> and access the platform <a href="https://aic.onelab.eu/">https://aic.onelab.eu</a>.</p><p>We would be more than glad to have you onboard !</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8571b06c424d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/alter-way-innovation/announcing-aic-2-0-an-open-source-android-cloud-emulator-8571b06c424d">Announcing : AiC 2.0, an Open Source Android cloud emulator</a> was originally published in <a href="https://medium.com/alter-way-innovation">alter-way-innovation</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>