The Nth Region Project: An Open Retrospective
summary
New Relic rebuilds an EU region in IBM Frankfurt to run ~500 services from ~50 teams, converting services to containers, standardizing CoreOS, Vault for secrets, and Terraform for provisioning; service discovery becomes dependency-injected environment variables so static analysis can map dependencies; three test buildouts uncover networking and rollout thrash, a pilot with a few customers proves reliability, engineering hours for region build drop by an order of magnitude, and the goal of one small team per new region is nearer but not yet achieved.
AI-written summary · about
chapters (11)
- 0:00Start
- 1:08New Relic Teams with IBM to Expand in Europe, Speed Cloud Adoption
- 2:46600 MILLION EVENTS PER MINUTE
- 4:54MIGRATIONS AND DISASTER RECOVERY
- 10:14Interfaces separate the contract from the implementation
- 11:21Encapsulation lets small teams act independently
- 11:36HIGH LEVERAGE INTERFACES
- 19:00The Goldilocks balance: Choose the right work
- 20:34Quick ramp-ups How do you prioritize work?
- 26:40Steel thread: Validate a design using a subproject that tests it thoroughly.
- 35:5995% SERVICES IN CONTAINER FABRIC
related talks
37mMoving from 1 to N regions: an open retrospective
23mDay-3-Operations: New Region and New Services
27mBuilding Multi-region CI/CD Pipelines with a DevOps Mindset
53mBuilding Portable Infrastructure for New Relic
29mMulti Region Cloud Architecture for Data Residency Compliance
11mBuilding a system that never stops: New Relic at scale
51mAchieving Resiliency With A Multi-Cloud Or Multi-Region Approach
35mMulti Region Terraform Deployments