We're happy to have two talks about cutting edge technologies that have been getting a lot of attention lately (AWS' Firecracker and BPF) and their potential for integration with Kubernetes.

In the first talk, Dongsu will look at what Firecracker is and does. We'll then have a look at the potential for integration with container runtimes.
In the second talk, Alban will demonstrate the potential for BPF tracing features in Kubernetes via the project kubectl-trace.

This meetup will serve as a warmup to the FOSDEM talks that will take place the following weekend in Brussels.

Snacks & drinks will be provided by the CNCF and Kinvolk.

---------------------------------------------

SCHEDULE:

[18:30- 19:00] Arrive and mingle

[19:00 - 19:30] Dongsu Park, "Firecracker, should it work only with a single runtime?"

[20:00 - 20:45] Alban Crequy, "'Introducing kubectl-trace"

[20:45 - 21:15] Mingle time

---------------------------------------------

TALK DETAILS:

"Firecracker, should it work only with a single runtime?", Dongsu Park

Abstract: Firecracker, an open-source project published by Amazon, opens up a lot of interesting perspective to the container ecosystem. For the diversity of Kubernetes ecosystem, we should also think about how we should get the new shape of containers integrated with all other high-level container runtimes, such as cri-o or rktlet.

Firecracker, an open-source project published by Amazon, opens up a lot of interesting perspective to the container ecosystem. It opens up use cases between containers and VMs, making use of KVM without relying on heavyweight Qemu instances. Though its integration with the Kubernetes ecosystem has only started its first step, only with containerd. For the diversity of Kubernetes ecosystem, we should also think about how we should get the new shape of containers integrated with all other high-level container runtimes, such as cri-o or rktlet.

In this talk, I would like to talk about alternative ways of integration for getting firecracker integrated with multiple container runtimes. A proof-of-concept for the concept can be also presented during the talk.

-----

"'Introducing kubectl-trace", Alban Crequy

Abstract: I will show how to use BPFtrace to trace both function calls in your programs and kernel functions. Then, I will show how such a low-level tracing tool can be used in high-level environments like Kubernetes clusters with kubectl-trace.

kubectl-trace (https://github.com/fntlnz/kubectl-trace) schedules bpftrace programs on your kubernetes cluster using the kubectl.

BPFtrace (https://github.com/iovisor/bpftrace) is a new tracing tool announced in October 2018. You can write one-liner commands in your shell using a high-level tracing language inspired by awk and C, and predecessor tracers such as DTrace and SystemTap. Behind the scenes, it compiles this high-level language with LLVM into eBPF bytecode (extended Berkeley Packet Filter) available in recent Linux kernels. It allows you to trace different layers of the system: kernel dynamic tracing (kprobes), user-level dynamic tracing (uprobes), and tracepoints.

It can be used to trace containers, for example running on Kubernetes, the most popular container orchestration tool. BPFtrace is used as a base tool by kubectl-trace to build more complex tracing tools on Kubernetes.
---------------------------------------------

SPEAKER BIO: Coming soon