Serverless Java hacks: Think different and leave old habits...
summary
Explains concrete serverless Java practices: prefer static global cache for short lived reuse, avoid heavy frameworks that create pools and instead avoid connection and thread pools by keeping one connection/thread per invocation, use event driven flows to prevent paying while waiting, target startup optimizations with GraalVM native compilation and AWS SnapStart or CRaC, consider Quarkus and native HTTP clients, use Java CDK or Pulumi for infra, and design for bursts like 10,000 parallel invokes.
AI-written summary · about
chapters (11)
- 0:00Introduction and speaker intro
- 1:57What is serverless computing
- 4:09Key serverless Java challenges
- 8:21Handling state in serverless
- 11:11Static variables as cache
- 14:27Event-driven approach benefits
- 16:59Avoid resource pooling
- 17:20Speed matters in serverless
- 19:32Infrastructure as Java code
- 20:57Cold starts and JVM optimizations
- 22:43Q&A session
related talks
53mAdapting Java for the Serverless world
31mServerless Java in Action: Cloud Agnostic Design Patterns and Tips
53m"Serverless" Is What J2EE Was Meant To Be
01h 00mAdapting Java for the Serverless world
51mAdapting Java for the Serverless world!
01h 03mKeynote: Why The Future of Java in the Cloud is Serverless
36mServerless Java
52mContainerless, Bloatless, YAML-less, Serverless Java #slideless