Effortless Profiling on Kubernetes

Conference: KubeCon + CloudNativeCon North America 2021

2021-10-13

Authors: Eden Federman

Summary

Profiling is the act of analyzing the performance of applications in order to improve poorly performing sections of code
Flame graphs are a popular way to visualize a profile
The challenges of profiling include overhead and modifying code
Cuba City Flame is a tool that aims to make profiling effortless by removing the need to do code modifications and by doing profiling without having to do a deployment
The future of profiling includes ephemera containers, ebpf, and continuous profiling tools

Cuba City Flame is a tool that simplifies the process of profiling by removing the need to modify code and by doing profiling without having to do a deployment. This tool saves users from having to learn which profiler to use and the specific API details. It also eliminates the need to copy over files and restart applications. Cuba City Flame is a cube ctl plugin that can be easily installed via crew. It makes profiling much more accessible and saves time and effort.

Abstract

Finding performance issues in applications is a hard problem. It's even harder when those applications run on Kubernetes. The needed profiling tools are usually not included in the container image and require a new deployment in order to be enabled. This session covers the best practices and the different tools available to profile Java, Go, Python, and Ruby applications.

Materials:

Tags: