KServe Triton Latency
KServe Triton Latency Dashboard
Latency metrics for pre/post-process, predict, and explain steps for Triton Inference Service with KServe & total number of requests. Requires inferenceservices to use KServe >= v0.10. See KServe observability docs for more info.
Data source config
Collector config:
Upload an updated version of an exported dashboard.json file from Grafana
Revision | Description | Created | |
---|---|---|---|
Download |