Serverless Systems That Won't Burn a Hole in Your Pocket
summary
Claims concrete savings and tools: switching to Arm yields 20% cost saving on Arm, provisioned concurrency—when sized to base load—can be 60% cheaper with provisioned concurrency, and rebalancing memory can make a 128MB job eight times faster at nearly the same cost. Practical fixes shown and measured: move heavy setup into init to cut cold-start invoke time from ~888ms to ~3ms, use Lambda Power Tuning and Compute Optimizer to pick memory, replace busy-wait Lambdas with Step Functions or EventBridge to eliminate idle billing.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 4:20Cost matters in serverless
- 5:22Lambda billing explained
- 11:11Memory latency vs throughput
- 15:23Optimizing cold starts
- 22:31Optimal memory allocation
- 25:18Reducing idle time
- 29:09Offloading requests
- 31:15Avoiding unnecessary invocations
- 33:21Direct integrations
- 37:19Cost optimization takeaways
- 39:06Q&A
related talks
42mServerless Systems That Won't Burn a Hole in Your Pocket
58mServerless: Five Key Things you need to Know
10mEnergy-Efficient and Cost-Optimized Serverless Computing
27mServerless Performance on a Budget
35mEfficiency of Serverless Architectures with AWS Solutions
51mAdapting Java for the Serverless world!
47mMoney-saving tips for the frugal serverless developer
43mBest Practices for Serverless Developers