How to Train your Jenkin to Use Your Resources Effectively and Get More Value
summary
A legacy model issues one Jenkins VM per developer but only 20% of VMs are used and active instances sit idle for less than 25 minutes per day, wasting millions. The team replaces per-VM Jenkins with a Mesos/Marathon cluster plus a Jenkins–Mesos plugin, running lightweight Jenkins masters (about 0.2 CPU and 256 MB) and spawning build slaves on demand; result: provisioning drops from minutes to seconds, higher density (many Jenkins per host), HA and measurable cost savings.
AI-written summary · about
chapters (6)
- 0:00Introduction to CI at eBay
- 1:22Legacy system challenges
- 2:40Transitioning to Apache Mesos
- 7:10New architecture design
- 9:39Results and performance
- 12:04Q&A and closing
related talks
35mHow to Train Jenkins?- 35mHow to Train Jenkins?
42mSeven Habits of Highly Effective Jenkins Users- 42mSeven Habits of Highly Effective Jenkins Users
55mContinuous Integration on Steroids
35mJenkins Scaling and Organization for an Efficient CI
47mJenkins in the Enterprise: Building Resilient CI Infrastructure
38mOperating Jenkins at Scale