Jenkins World 2017

How many Jenkins pipelines can I run at once? What limits the speed and scale? How can I make my pipelines run faster?

To answer these questions and more, I’ll draw on a mix of large-scale user cases, testing environments, theory and micro-benchmarks.

This talk will bring together several areas of interest to users trying to run very large Jenkins instances with Jenkins Pipeline…or very small, resource-constrained instances!

Topics for discussion:
- Resource use: memory, storage I/O and CPU use
- Specific bottlenecks: code vs. framework vs. data persistence
- Hardware demands and systems tuning
- What makes things faster?

To answer these speed and scalability questions, we will draw from four different sets of data:
- Field observations assisting with some of the largest-scale Pipeline users
- Data collected from a pipeline scalability lab environment
- Programming work on pipeline internals, to cover theory and implementations
- Microbenchmarks of use-cases and specific implementation components.

Understanding these concepts will help you estimate resource needs for instances AND reduce them.