-
blue : vertx 4.3.4
-
red : vertx 4.3.4 + async await
-
orange : vertx 4.3.4 + async await using the event-loop thread as carrier thread
-
plaintext performance look dramatic because this benchmark uses 16 level pipelining. The current implementation of vertx virtual thread is not optimized for this and will flush each request, whereas the default version will flush every 16 requests (after the buffer containtaing the 16 pipelined request has been processed).
-
other benchmark simply shows that using virtual thread has a cost
Last active
February 1, 2024 03:09
-
-
Save vietj/fe9f886d489853ab07111ba5715b13ee to your computer and use it in GitHub Desktop.
Vert.x virtual thread benchmark
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Is there updated benchmark using vertx 4.5 and virtual threads? If virtual thread does not bring performance improvement, is't encouraged to use it in vertx 4.5?