Software Development Analytics Jesus M. Gonzalez-Barahona with - PowerPoint PPT Presentation

Analytics with GrimoireLab Software Development Analytics Jesus M. Gonzalez-Barahona with GrimoireLab A bit of context Dealing with dynamic complexity Data sources GrimoireLab Jesus M. Gonzalez-Barahona Cauldron Alpha Universidad Rey Juan Carlos Case studies @jgbarah http://github.com/jgbarah/presentations Activity Remaining code Performance Demographics Intl. Summer School on Visual Soft. Analytics Diversity Final remarks Leipzig (Germany), September 23rd 2019 Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 1 / 76

Analytics with GrimoireLab Jesus M. Gonzalez-Barahona A bit of context It is difficult to improve Dealing with dynamic complexity Data sources if you cannot measure GrimoireLab and track your improvement Cauldron Alpha Case studies Activity Remaining code Performance Demographics Diversity Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 2 / 76

Analytics with Our plan today GrimoireLab Jesus M. 1 A bit of context Gonzalez-Barahona Dealing with dynamic complexity A bit of context 2 Data sources Dealing with dynamic complexity 3 GrimoireLab Data sources 4 Cauldron Alpha GrimoireLab 5 Case studies Cauldron Alpha Activity Case studies Remaining code Activity Remaining code Performance Performance Demographics Demographics Diversity Final remarks Diversity 6 Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 3 / 76

A bit of context Analytics with GrimoireLab Jesus M. Gonzalez-Barahona A bit of A bit of context Dealing with dynamic complexity Data sources context GrimoireLab Cauldron Alpha Case studies Activity Remaining code Performance Demographics Diversity Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 4 / 76

A bit of context Analytics with Me and my two hats GrimoireLab Jesus M. Gonzalez-Barahona Uni Rey Juan Carlos: A bit of context • Understanding free, open source Dealing with dynamic complexity Data sources software GrimoireLab Cauldron Alpha • Data analytics approach Case studies • Data visualization in XR Activity Remaining code Performance Demographics Diversity http://gsyc.es/jgb Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 5 / 76

A bit of context Analytics with Me and my two hats GrimoireLab Jesus M. Gonzalez-Barahona Bitergia: A bit of context • From research to the real world Dealing with dynamic complexity Data sources • Understanding software GrimoireLab Cauldron Alpha development Case studies • Data analytics approach Activity Remaining code Performance Demographics Diversity http://bitergia.com Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 6 / 76

A bit of context Analytics with Recommendations GrimoireLab Jesus M. Gonzalez-Barahona • Open your laptop A bit of context • Download the slides (they have links) Dealing with dynamic complexity Data sources • Visit Alpha.Cauldron.io and produce your GrimoireLab own dashboard Cauldron Alpha • Play with the dashboards Case studies Activity • Understand the interpretations behind the Remaining code Performance Demographics Diversity numbers Final remarks https:/alpha.cauldron.io Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 7 / 76

A bit of context Analytics with Cauldron Alpha GrimoireLab Jesus M. Gonzalez-Barahona A bit of context Dealing with dynamic complexity Data sources GrimoireLab Cauldron Alpha Case studies Activity Remaining code Performance Demographics Diversity Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 8 / 76

A bit of context Dealing with dynamic complexity Analytics with GrimoireLab Jesus M. Gonzalez-Barahona A bit of context A bit of context Dealing with dynamic complexity Data sources Dealing with dynamic complexity GrimoireLab Cauldron Alpha Case studies Activity Remaining code Performance Demographics Diversity Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 9 / 76

A bit of context Dealing with dynamic complexity Development projects may be large and complex Analytics with GrimoireLab Jesus M. Gonzalez-Barahona A bit of context Dealing with dynamic complexity Data sources GrimoireLab Cauldron Alpha Case studies Activity Remaining code Performance Demographics Diversity Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 10 / 76

A bit of context Dealing with dynamic complexity Analytics with Projects may be large and complex... GrimoireLab and dynamic Jesus M. Gonzalez-Barahona It’s difficult to... A bit of context Dealing with dynamic complexity • ...track what’s happening Data sources • ...understand why it’s happening GrimoireLab Cauldron Alpha • ...react quickly Case studies • ...evaluate results of reaction Activity Remaining code Performance Demographics Diversity If data is available Final remarks analytics may come to the rescue Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 11 / 76

A bit of context Dealing with dynamic complexity Analytics with A continuous process GrimoireLab Jesus M. Gonzalez-Barahona Figure out your interest A bit of context Find out available data Dealing with dynamic complexity Data sources Define key parameters GrimoireLab Monitor, understand, detect deviations Cauldron Alpha Act to correct, improve Case studies Activity Remaining code Track results Performance Demographics Diversity Final remarks Measure → Monitor → Act Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 12 / 76

A bit of context Dealing with dynamic complexity Analytics with A continuous process GrimoireLab Jesus M. Gonzalez-Barahona Case example: Overall development activity A bit of context Dealing with dynamic complexity Interest: activity Data sources Data: changes to code, tickets GrimoireLab Cauldron Alpha Parameters: commits, tickets closed Case studies Monitoring: charts, numbers Activity Remaining code Performance Demographics Observation: numbers declining Diversity Final remarks Action: allocate more developer effort Track results... Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 13 / 76

Data sources Analytics with GrimoireLab Jesus M. Gonzalez-Barahona A bit of context Data sources Dealing with dynamic complexity Data sources GrimoireLab Cauldron Alpha Case studies Activity Remaining code Performance Demographics Diversity Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 14 / 76

Data sources Analytics with Repositories, repositories... GrimoireLab Jesus M. Gonzalez-Barahona A bit of context Dealing with dynamic complexity Data sources GrimoireLab Cauldron Alpha Case studies Activity Remaining code Performance Demographics Diversity Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 15 / 76

Data sources Analytics with Source code management GrimoireLab Jesus M. Gonzalez-Barahona • Client/server: CVS, Subversion A bit of context Dealing with dynamic complexity • Decentralized: git, Mercurial, Bazaar, etc. Data sources GrimoireLab • Most of them accessible through git... Cauldron Alpha (with some problems) Case studies Activity Remaining code • Can be integrated with other tools: Performance Demographics Diversity Gerrit, GitHub, GitLab, etc. Final remarks Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 16 / 76

Data sources Analytics with Issue tracking GrimoireLab Jesus M. Gonzalez-Barahona Many different systems: A bit of context • Bugzilla Dealing with dynamic complexity Data sources • Jira GrimoireLab • GitHub issues Cauldron Alpha • GitLab Issues Case studies Activity Remaining code • Phabricator Performance Demographics Diversity • RedMine... Final remarks Each with a different model, data, operations... Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 17 / 76

Data sources Analytics with Code review GrimoireLab Jesus M. Gonzalez-Barahona Usually: peer review pre-merge review A bit of context Different methods: Dealing with dynamic complexity Data sources • Mailing lists (eg: Linux) GrimoireLab • Gerrit (eg: OpenStack) Cauldron Alpha • GitHub pull requests (eg: ElasticSearch) Case studies Activity • GitLab merge requests (eg: GNOME) Remaining code Performance Demographics • or even Jira, Bugzilla... Diversity Final remarks Much of the control on the software lies here Jesus M. Gonzalez-Barahona (URJC) Analytics with GrimoireLab Visual Software Analytics 18 / 76