[CNCF LFX Proposal] HAMi: HAMi GPU Observability: Metrics and Dashboards
- Dominant language
- JavaScript
- Stars
- 3.1k
- Forks
- 816
- Avg merge
- 12h 32m
- Merged PRs (30d)
- 8
Description
### CNCF Project
HAMi
### Term
2026 Term 3 (Sep-Nov)
### Program Name
HAMi GPU Observability: Metrics and Dashboards
### Program Description
## Description
Improve the usefulness, consistency, and documentation of HAMi GPU observability for operators running shared GPU workloads. The program will focus on practical metrics and dashboards, while keeping tracing work bounded to a design and proof of concept for one mentor-approved control-plane path.
## Expected outcomes
- A gap analysis of current HAMi metrics, exporters, labels, dashboards, and documentation.
- Maintainer-reviewed improvements to a focused set of Prometheus metrics and labels across the HAMi device plugin and scheduler extension.
- A versioned Grafana dashboard covering selected allocation, utilization, memory, failure, and scheduling signals.
- Documentation of metric semantics, units, labels, cardinality, and example PromQL.
- An OpenTelemetry trace design and bounded proof of concept for one selected workflow.
- Practical alerting or SLO guidance grounded in the delivered telemetry.
- At least one reviewed and merged contribution involving metrics, dashboards, tests, or documentation.
### Technologies
Prometheus, Grafana, OpenTelemetry, Kubernetes, Go, HAMi
### Skills same as Technologies?
- [ ] Yes, the required skills are the same as the technologies listed above.
### Required/Desirable Skills
Metrics instrumentation, PromQL, dashboard design, observability, Kubernetes debugging, HAMi internals
### Mentors
Mesut Oezdil | @mesutoezdil | mesudozdil@gmail.com | mesutoezdil777
Reza Jelveh | @fishman | reza.jelveh@gmail.com | softsage
Jimmy Song | @rootsongjc | rootsongjc@gmail.com | jimmysong
### Upstream Issue URL
https://github.com/Project-HAMi/HAMi/issues/2126
### Application Prerequisites
- [x] Resume
- [x] Cover Letter
- [ ] School Enrollment Verification
- [ ] Participation Permission from school or employer
- [ ] Coding Challenge
- [ ] Custom Prerequisite (fill in details below)
### Coding Challenge URL
_No response_
### Custom Prerequisite Name
_No response_
### Custom Prerequisite Description
_No response_
### Custom Prerequisite — File Upload
- [ ] Yes — completion of this task requires the mentee to submit a file.
---
**LFX program:** [CNCF - HAMi: HAMi GPU Observability: Metrics and Dashboards (2026 Term 3)](https://mentorship.lfx.linuxfoundation.org/project/dd302799-03ec-4184-b289-4d59a41fe7ed)
Contributor guide
Research direction
Start with upstream issue #2126 and HAMi’s current metrics, exporters, dashboards, and documentation. Map gaps across the device plugin and scheduler extension, then confirm the focused Prometheus and Grafana scope with mentors. Done means reviewed metrics and labels, a versioned dashboard, metric documentation, and a bounded OpenTelemetry proof of concept with tests or merged contributions.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go, grafana, kubernetes, prometheus
- Domain
- documentation, observability-sre
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100