Can enterprises use their existing monitoring tools to keep track of containerized workloads? This episode of The New Stack Makers podcast features an interview with Canturk Isci, IBM research staff member. Isci discusses how monitoring practices are evolving through the principles of operational visibility. TNS technical editor Benjamin Ball interviewed Isci for the final installment of our ebook series on the Docker and container ecosystem, the one focusing on monitoring containers.
The interview can also be heard on YouTube.
When it comes to monitoring, many developers find themselves at a crossroads when building out their containerized infrastructure, wondering if their existing monitoring solution can cope. Isci noted that while the jump from traditional systems to VMs was bearable, cracks eventually were visible when it came to maintenance of one’s infrastructure. “You are basically trying to build a whole city around one store that tries to serve one thing and basically the other thing I’m going to have here is my fire station, here’s my police, and all of it. Basically if you want to build that store, you have to build all of that, otherwise, you won’t be able to observe it,” said Isci.
Isci noted that while standard application performance monitoring (APM) providers are still strong presences within this section of the container ecosystem, more grassroots options are emerging that bill their solutions as being hyper-specific to managing every aspect of the container lifecycle.
“Under the covers, though, if you look at core principles and build solutions that leverage that, then you’ll find a lot of interesting opportunity. You can actually now understand how a group of containers work together, how they actually build from the same configuration management, and how they actually utilize resources,” noted Isci.
IBM sponsored this article.
Feature image: The interlocking machine that ran inside the Pennsylvania Railroad’s Harris Switch Tower in Harrisburg, Pennsylvania, under the Creative Commons Attribution 3.0 Unported license.