QCon10Years
Conference:March 7-9, 2016
Workshops:March 10-11, 2016
Sign up to be informed when registration opens for QCon London 2017 (3/6 - 3/10)

Presentation: Observe, Enhance, & Control: VMs to Containers

Track: 
Containers (in production)

Location:

  • Whittle, 3rd flr.

Duration

Duration: 
1:40pm - 2:30pm

Day of week:

  • Tuesday

More talks on:

Key Takeaways

  • Understand some of the challenges inherent when you move to an architecture that leverages containers.
  • Hear where we still need innovation in the container space to handle the scale containers offer.
  • Learn that while the high level problems remain the same, the implementation details of using containers is different that what you may have experienced with VMs.

Abstract

Operationally, many of the problems we're solving for containers have long been solved for VMs. However, these solutions don't quite fit the container model and require a new perspective and implementation. In this talk, we take a look at the theme of this track "Observe, Enhance, Control" and see how this related to VMs, what solution architectures worked there, and discuss why these architectures are no longer adequate in a containerized world.

We'll then look at how these same problems are being solved today, and look to the future of how we can expect similar problems to be solved as the container world matures.

Interview

Question: 
Can you explain your title a bit to me?
Answer: 
“Observe, Enhance, Control” is the theme of this track and is a good generalization of the goals of any application deployment and management. Sure, we could get more specific and add other topics like “security,” but I think Observe, Enhance, Control is a good generalization.
I believe this generalization applies to all current computing paradigms, and so I wanted to give a talk about how this applies to our model of thinking from VMs to containers.
Question: 
What’s the motivation for your talk?
Answer: 
At this point, operators and developers are very comfortable with VMs. As we work with containers, we quickly realize that a lot of the ways we work with VMs no longer apply to containers. Why is that? What do we do instead? I want to give this talk to help answer these questions.
Question: 
Can you give me an idea of a pattern that worked with VM’s, but just doesn’t hold for containers?
Answer: 
Manually handling placement of all deploys. Or security/identity of everything manually. There are a lot of examples that go towards the common thing I’m getting at: scale. Containers help realize scale and density that is difficult to manually manage. You can manually manage it but then you’ll be giving up a lot of the benefits of containers!
Question: 
The last sentence of your abstract stands out to me: “We'll then look at how these same problems are being solved today, and look to the future of how we can expect similar problems to be solved as the container world matures.” Can you elaborate?
Answer: 
If you break down “Observe, Enhance, Deploy” from VMs and compare to containers, there are very clear areas that we haven’t seen innovation from in the container space. I argue that we will and that we can predict what they’ll likely look like.

Mitchell Hashimoto Elsewhere

GitHub

Similar Talks

Multi-Host, Multi-Network Persistent Containers
@WillHillBet: Love failure & embrace the fall out
Hot code is faster code - addressing JVM warm-up
Automating Security at Slack
Nihilist’s Guide to Wrecking Humans and Systems
Bitcoin Security: 1/10th cent to a billion dollars
Building a Modern Security Engineering Team
Real-Time Fraud Detection with Graphs
Successful Go program design, 6 years on

Tracks

Covering innovative topics

Monday, 7 March

Tuesday, 8 March

Wednesday, 9 March

FULL SCHEDULE

Conference for Professional Software Developers

Follow QCon

Contact

Menu

QCons around the World

Qcon