CI at Uber scale: Lean, green and fast
This session will explore how Uber’s MergeQueue, an innovative scheduling system, ensures mainline stability and production readiness for every commit, even with thousands of changes submitted per hour across multiple monorepos.
At Uber's scale, maintaining an always-green mainline while processing hundreds of changes per hour from thousands of engineers posed a significant challenge. Dhruva Juloori will delve into the design and evolution of MergeQueue, Uber’s advanced scheduling system that guarantees mainline stability and production readiness with every commit. By utilizing speculative validation, build-time predictions, and sophisticated conflict resolution techniques, MergeQueue enhances efficiency, reduces CI bottlenecks, and optimizes resource usage. This system empowers thousands of engineers to ship code quickly and reliably, ensuring that every commit is production-ready and seamlessly integrated into Uber’s ecosystem.
CI at Uber scale: Lean, green and fast
This session will explore how Uber’s MergeQueue, an innovative scheduling system, ensures mainline stability and production readiness for every commit, even with thousands of changes submitted per hour across multiple monorepos.
Panelist

Panelist

Panelist

Moderator

Dhruva Juloori
Senior Software Engineer, Uber
At Uber's scale, maintaining an always-green mainline while processing hundreds of changes per hour from thousands of engineers posed a significant challenge. Dhruva Juloori will delve into the design and evolution of MergeQueue, Uber’s advanced scheduling system that guarantees mainline stability and production readiness with every commit. By utilizing speculative validation, build-time predictions, and sophisticated conflict resolution techniques, MergeQueue enhances efficiency, reduces CI bottlenecks, and optimizes resource usage. This system empowers thousands of engineers to ship code quickly and reliably, ensuring that every commit is production-ready and seamlessly integrated into Uber’s ecosystem.
CI at Uber scale: Lean, green and fast
This session will explore how Uber’s MergeQueue, an innovative scheduling system, ensures mainline stability and production readiness for every commit, even with thousands of changes submitted per hour across multiple monorepos.
At Uber's scale, maintaining an always-green mainline while processing hundreds of changes per hour from thousands of engineers posed a significant challenge. Dhruva Juloori will delve into the design and evolution of MergeQueue, Uber’s advanced scheduling system that guarantees mainline stability and production readiness with every commit. By utilizing speculative validation, build-time predictions, and sophisticated conflict resolution techniques, MergeQueue enhances efficiency, reduces CI bottlenecks, and optimizes resource usage. This system empowers thousands of engineers to ship code quickly and reliably, ensuring that every commit is production-ready and seamlessly integrated into Uber’s ecosystem.
CI at Uber scale: Lean, green and fast
This session will explore how Uber’s MergeQueue, an innovative scheduling system, ensures mainline stability and production readiness for every commit, even with thousands of changes submitted per hour across multiple monorepos.
Panelist

Panelist

Panelist

Host

Dhruva Juloori
Senior Software Engineer, Uber
Sign up now

