Automation in Distributed Systems
Automating a test case/suite is one of the key challenges in Distributed Computing Systems, apart from deployment and monitoring. In this crisp talk I will briefly speak about different tools avaiable, and pros and cons of each. And conclude why we ended up using rpyc - A python based rpc server.
Remote deployment, monitoring, and automation are the three major challenges for people working in distributed computing systems. This becomes harder when the number of systems are scaled up to hundreds or even thousands of nodes.
I will first brief about available tools for each of those challenges, like python-fabric, puppet, STAF, and python-rpyc. I will briefly compare the pros and cons of each of those tools and use cases they are best suitable for.
Then I will speak about how we were able to solve most of our test suite automation problems using rpyc. Explain how rpyc can be used for deployment, monitoring, and testing a distributed file system.
If time permits I will give a small demo, without doing any code walk through.
Basic linux and python programming.
I am an automation engineer at Red Hat Inc. I write and automate scripts to test GlusterFS, a distirbuted file system.
Ealier I was an engineer at Aricent, working on testing protocol stack in IMS communication systems.
I deal with distributed systems as part of daily job.