intel / cloud-native-ai-pipeline

AI cloud native pipeline for confidential and sustainable computing
https://intel.github.io/cloud-native-ai-pipeline/
Apache License 2.0
37 stars 15 forks source link

ui: add latency data to the echarts #132

Closed Hulongyin closed 1 year ago

Hulongyin commented 1 year ago

This PR is to fix #131 , add latency data to the echarts on SPA.

spa_latency

The whole bar represents the end to end latency, the bottom half of the bar represents the stream to inference (end) latency, and the top half of the bar represents the inference (end) to ui latency. The chart update every second (so the latency is average over 1s) and has up to 5 bars.