https://github.com/ray-project/serve_workloads
https://github.com/ray-project/serve_workloads
Last synced: 9 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/ray-project/serve_workloads
- Owner: ray-project
- Created: 2023-03-08T22:19:32.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2024-08-12T21:29:04.000Z (almost 2 years ago)
- Last Synced: 2025-03-26T18:51:56.074Z (over 1 year ago)
- Language: Python
- Size: 237 KB
- Stars: 2
- Watchers: 3
- Forks: 5
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Serve Workloads
This repo contains long-running Serve workloads that are meant to test important
Serve features such as autoscaling, deployment graphs, and high availability.
These workloads are generally split into two applications– a `Pinger` and a
`Receiver`. The `Pinger` app sends queries to the `Receiver` app and emits logs
to track success rates and latency. The `Receiver` tests particular Serve
features such as autoscaling or deployment graphs.
The `Pinger`'s metrics can be viewed in a Grafana dashboard, letting users
monitor the workload over long periods of time.