Repository navigation
chore: promote staging to main (benchmarks, open-loop Poisson, scenario diagrams, agentcore-demo) - #72
Merged
Conversation
…uster
- gateway-deployment.yaml takes the CA ClusterIssuer from state
({{.BNK_CA_ISSUER}}) instead of the Sydney cluster's name, so the TLS
certificate issues on a cluster with another name
- the scripts use the Gateway name the manifests fix
(bnk-agentcore-demo-gateway) instead of <cluster>-gateway
- rebuild.sh merges this cluster's OIDC provider into the shipper role's
trust policy and keeps the other live clusters' statements
- forge target registration: when Forge already holds the target name for
another cluster (names are global in Forge), register this cluster's
endpoint as <name>-<cluster> instead of silently reusing the other
cluster's target
Verified on agent-core-singapore: certificate Ready with the rendered
issuer; forge scan --register-targets created target 7
mcp-default-mcp-financial-route-agent-core-singapore on cluster 21 where
it previously returned Sydney's target 5.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
fix(agentcore-demo): run on any cluster name; own Forge target per cluster
… on auth rejection
…m KV relay sidecar - Support open-loop Poisson scenarios (poisson-rate, poisson-prefix, poisson-agentic) with dynamic request rates and automatic dataset size calculation. - Embed ZMQ KV relay sidecar (sim_kv_relay.py) in the ai-inference scenario so both F5 EPP and llm-d receive real KV cache events from vLLM/sim pods. - Include Gateway API Inference Extension v1.5.0 CRD manifests (InferencePool). - Discover served model names from the target inference deployment and pass synthetic tags to Forge. - Support agent daemon open-loop dispatch with agent-bound token handling. - Add SageMaker intent inference model and synthetic spec helpers.
…ck test - Upgrade github.com/containerd/containerd to v1.7.36 to resolve GO-2026-6444 and GO-2026-6597. - Add Seqs to cspell.json allowed words list for maxNumSeqs in docs/SCENARIOS.md. - Use polling with deadline in TestBenchmarkDaemon_EndToEndMock to eliminate flaky sleep timing on CI runners.
feat(benchmark): open-loop Poisson scenarios, daemon dispatch, and sim KV relay sidecar
docs(benchmarks): add benchmarking guide, scenario diagrams, and stakeholder landing page
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.
Promotes staging to main. Contains:
• #68
fix(agentcore-demo): run on any cluster name; own Forge target per cluster.• #69 / #71
fix(forge): use an agent-bound token for the benchmark agent, annotate false-positives for gosec, and fail on auth rejection.• #70
feat(benchmark): open-loop Poisson scenarios (poisson-rate,poisson-prefix,poisson-agentic), daemonized scenario dispatch, sim KV relay sidecar (sim_kv_relay.py), Gateway API Inference Extension CRDs (v1.5.0InferencePool), served model auto-discovery, and SageMaker IRSA OIDC phase handling.• #71
docs(benchmarks): comprehensive AI inference benchmarking guide (docs/BENCHMARKS.md), dedicated architecture diagrams for all 15 validation scenarios and 5 curated walkthrough demos indocs/SCENARIOS.md, and complete README landing page overhaul for enterprise customers, SEs, and sales.All changes have passed CI (100% green across all 13 check suites).