StreamdataIO at DevoxxFR | API Streaming

All the streamdata.io team was at DevoxxFR. It was a great event! Many thanks to the organizers who did a great job!

Most of the time, as we did for QConLondon, people write blog posts about the conferences they attended (and lots were excellent at DevoxxFR as the keynote of Eric Filiol and the one of Rodolphe Gelin and, thanks to the DevoxxFR Team, the videos will be soon available on Parleys. So, rush on them once they will be published!) But for once, I will write about what we have seen on the sponsors booths.

There were quite lots of booths (compared to QConLondon). Some of well-known companies such as JFrog with their excellent Artifactory and Bintray or JetBrain with all the IntelliJ and friends stacks, some of big well known companies such as Google and IBM and some of less known companies…

Two of tools held our attention during DevoxxFr: Coverity and Logmatic.io.

Coverity is a code analysis and testing platform. From what I’ve understood, it can perform a static code analysis to detect potential defects and security vulnerabilities like many other tools. But it’s also a testing tool for QA people. It can perform tests coverage analysis and show which portions of code are covered by the tests. And when a piece of code changes after a commit, Coverity can reveal to the QA people which tests are impacted and must be run in priority to ensure there are no regressions or bugs. But the killer feature we saw, was the ability of Coverity to trace the code coverage from the web interface of a webapp to the server code. This was illustrated by a demonstration on a Java web application: a tester records a functional test by clicking on the web interface of the application (as he could have done with the Selenium or Gatling recorder) and Coverity is able to trace which server methods have been called through the UI (thanks to an agent run on the server side). Really amazing! And there are probably other gems we haven’t been demonstrated to. Very impressive tool for quality code insurance. Last words: Coverity works for Java, C/C++ and C# code. Javascript seems to be the next target. If you need a tool for quality insurance, it may worth having a look at it.

The second tool I enjoyed discovering at DevoxxFr was Logmatic.io. If you want to aggregate logs information from different servers and various sources (machine logs like cpu and memory consumption, application logs like jvm consumption or business metrics, etc.) and analyze the results, then Logmatic.io does the job. And it seems to do it pretty well with a nice web interface that aggregates the log information and enables the user to filter, aggregate and compute metrics based on the information of the different logs. You have lots of predefined filters and can create your own. You can also gather your information according to tags or categories you have formerly defined. Of course, there are also a various predefined log extractors (cpu and memory with syslog, apache logs, etc.) but you can also define your own “à la Logstash” style. Naturally, several types of nice charts are available to display the data and metrics. Add to this, the ability to define and receive alerts according your own criteria and you get a tool that looks like fantastic! From a pratical point of view, Logmatic.io can discuss with various technologies. We can cite syslog for machine level and logback and bro for the application level but this list is not exhaustive and there are lots more “natively” supported by Logmatic.io (and I guess it should be quite easy to plug your own log producer tool if it is not one that is “naturally” supported by Logmatic.io). The documentation looks like rich and describes for every technology how to feed Logmatic.io with logs. Another good point for them.

I guess the first targeted people are undoubtedly DevOps guys for all the monitoring goodies Logmatic.io brings. But I have also been demonstrated that marketing people can benefit from this platform by extracting relevant business information (such as the history of the number of users during the time, any business metrics you can add to your application, etc.). Possibilities of use look like vast. Several of my teammates went to this booth before me and all were seduced by this solution. It seems to be an ELK (Elastisearch – Logstash – Kibana) but power 10 (note : they use Elastisearch, so it may be more accurate to say it looks like a LK power 10 :)). Just one regret about Logmatic.io: the booth at DevoxxFR was really small and probably, lots of attendees have miss it. That’s a pity because I guess that lots of developers and ops would have been interesting in that SAAS platform… A tool-in action at DevoxxFR would have probably interested more than one of us too. Hopefully next time.