Being Creative: Tales from the Trenches
summary
Covers real-world Temporal misuses and fixes with concrete numbers and knobs: explains why replay is the baseline (not a slow path), shows how huge namespaces reach billions of open workflows, and demonstrates practical fixes — use continue as new, child workflows or the server batching API to trim history, tune workflow-task concurrency and cache size to avoid replay storms, and beware codecs causing KMS request spikes. Useful if deciding whether to redesign for durability vs. assumed linear scalability.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:09Creative uses of Temporal
- 3:39Understanding Temporal's internals
- 7:50Scalability and performance limits
- 12:32Workflow execution surprises
- 16:31Replay model and mental models
- 18:16Workflow cache optimization
- 19:42Reducing replay costs
- 24:58Workflow worker configuration
- 26:21Minimizing replay frequency
- 28:51Workflow cache sizing
- 30:21Conclusion and Q&A
related talks
26mMicroservices Madness: Application And Infrastructure Automation Using Temporal
49mHow to get smarter while doing your job
43mHow to Innovate in a Legacy
18mCode That Can't Fail
46mBlink and You’ll Miss It: Stories From Discovering Most Annoying Bugs and Test Flakes
01h 02mTools for simplified large scale systems
29mMaking tech boring to keep data exciting
36m"Engineering to embrace change" Or "How clumsy engineering can lead to powerful research – Part II"