Add Azure AKS kubernetes benchmarks for Thunder 1.0.0-alpha - #81
Open
UdeshAthukorala wants to merge 1 commit into
Open
UdeshAthukorala wants to merge 1 commit into
UdeshAthukorala wants to merge 1 commit into
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 (kubernetes) benchmark results for Thunder 1.0.0-alpha, following the same layout as
benchmarks/v0.40.0/kubernetes/.Contents
readme.md- summary table and per-run detail with K8s specsresults-1000conc.csv- raw summary for the 1000 concurrent user runresults-10000conc.csv- raw summary for the 10000 concurrent user runNotes for reviewers
The 10000 concurrent user run is not a valid benchmark and is marked as such in both the readme and the CSV. Error rates were 93.39% (client credentials), 98-100% (all five authorization code steps) and 99.85% (user authentication). The cause is load generator saturation rather than a Thunder limit: the single JMeter client (
Standard_F8s_v2, 8 vCore) reached a load average of 125.91, and 9963 of 9968 client sockets were inTIME_WAITat the end of the client credentials scenario. The cluster was also undersized at that level, with 2 nodes and Thunder capped at 10 pods of 1.5 vCore. It is included for traceability only and is excluded from the summary table.These runs used
USE_DELAYS=true, so authorization code throughput is bound by the 6000ms +/- 2000ms Gaussian think-time timer in the JMX rather than by the server. Client credentials and user authentication have no such timer. This differs from the v0.40.0 runs, which usedUSE_DELAYS=false, so the two sets are not directly comparable.The K8s specs are sourced from the committed pipeline defaults (
kubernetes/azure/devops-pipelines/variables.yaml,terraform/conf/thunder.conf.tfvars,templates/install-nginx.yaml), since the result bundles carry no infrastructure manifest.