Add kubernetes test results of v1.0.0-beta2 at 10000 concurrency - #87
Open
UdeshAthukorala wants to merge 1 commit into
Open
UdeshAthukorala wants to merge 1 commit into
UdeshAthukorala wants to merge 1 commit into
Conversation
UdeshAthukorala
force-pushed
the
add-k8s-benchmarks-1.0.0-beta2
branch
from
August 14, 2026 10:39
8fc69c7 to
4e1da14
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds the Azure AKS result for Thunder 1.0.0-beta2 at 10000 concurrent users, in the same layout as
benchmarks/v0.40.0/kubernetes: areadme.mdplus one CSV per scenario.Run: pipeline build 1500, 2026-08-13, performance repo at ff7cb7e. All stores on Postgres with
in-memory caching. Each scenario ran 15m with the first 5m discarded as warm-up.
The Authorization Code Grant Type and User Authentication with Credentials scenarios ran in the
same suite at this concurrency, but their error rates were too high to report as benchmarks, so
they are excluded. The readme records that they ran and were left out.
One caveat is recorded in the readme rather than left for the reader to discover: this run was
load generator bound. The JMeter client load average was 69.59 on a 64 vCore runner. In steady
state all 10000 threads were active at a 13ms average response while sustaining only about 16500
req/sec, roughly 1.65 iterations per thread per second, so the client rather than Thunder set the
ceiling. The v0.40.0 Postgres run at this concurrency has the same problem (client load average
119.78 on 72 vCores), so the two figures are not a clean comparison of server capacity.
The 797MB
results.zipis not included, matching the v0.40.0 entry.