Description
Describe the feature
I'd like to be able to pass a predefined request to the AwsCrtAsyncHttpClient in order to have it "warm" each connection in the pool on startup as well as each time a connection is cleaned up to recreate and re-"warm" before putting it back into the pool.
Use Case
I have a latency sensitive application that depends on DynamoDB and during deployment as well as in periods of low traffic it experiences spikes in p99 and p100 latencies. When investigating it seems that the latency from the DynamoDB call recorded from the client side is much higher than the latency recorded by DynamoDB which leads me to believe the overhead of new connections is what is causing this issue. There is also this issue that was closed seemingly without a solution: #1340
Proposed Solution
No response
Other Information
I would be happy to work on this feature, but I'd like to understand if this approach would be acceptable as an addition to the library.
Acknowledgements
- I may be able to implement this feature request
- This feature might incur a breaking change
AWS Java SDK version used
2.x
JDK version used
java11
Operating System and version
MacOS v10.15.7