The Pipeline Framework

Appearance

AWS Lambda SnapStart (Operate) ​

This page covers operational guidance for running TPF applications on AWS Lambda with SnapStart.

Quarkus SnapStart Support ​

With Quarkus Lambda applications, SnapStart support is included by the Quarkus Lambda stack and enabled by default.

Relevant keys:

  • quarkus.snapstart.enabled (current)
  • quarkus.snapstart.enable (deprecated alias)

Disable only when you explicitly need to:

properties
quarkus.snapstart.enable=false

JVM Startup Tuning ​

For SnapStart, Quarkus recommends tiered compilation configuration via environment variable:

bash
JAVA_TOOL_OPTIONS="-XX:+TieredCompilation -XX:TieredStopAtLevel=1"

This setting can also be useful for regular (non-SnapStart) Lambda functions with cold-start sensitivity.

  1. Verify Lambda runtime, memory, and timeout match your pipeline latency profile.
  2. Track cold-start and init duration separately from request duration.
  3. Validate tracing/metrics exporters under constrained Lambda memory sizes.
  4. Keep native and JVM deployment lanes in CI for regression coverage.

Tracing on Lambda ​

For AWS X-Ray integration on Quarkus Lambda:

  • add io.quarkus:quarkus-amazon-lambda-xray

Then configure tracing/export settings according to your observability topology.