Eero Tamminen

Adapter does not react to its config changes

1

**What happened?**: Changed prometheus-adapter (custom metrics configMap) config. Geneated custom metrics did not change until adapter was restarted (by deleting its pod). **What did you expect to happen?**: Similarly to...

Incomplete plugin SecurityContexts

2

**Describe the bug** Plugins do not match k8s _baseline_ [security standards](https://kubernetes.io/docs/concepts/security/pod-security-standards/) (mainly due to host mounts), but that could be compensated somewhat by adopting relevant policies from the k8s _restricted_...

Update PrometheusAdapter config based on "MetricRules" objects

7

### Component(s) _No response_ ### What is missing? Please describe. Operator updating PrometheusAdapter config based on available `MetricRules` objects, similarly to how it already updates Prometheus config based on `PodMonitor`...

[Bug] doc links in application READMEs point to dirs instead of READMEs within them

1

### Priority P3-Medium ### OS type N/A ### Hardware type N/A ### Installation method - [ ] Pull docker images from hub.docker.com - [ ] Build docker images from source...

Improve DocSum file handling

3

## Description Use temporary file only when necessary, and use `aiofiles` own functionality for that. ## Issues `n/a`. ## Type of change - [x] Others (enhancement, documentation, validation, etc.) ##...

[BUG] TGI versions inconsistency / use of old TGI versions

10

Currently latest used TGI versions in this repo are v2.3.1 (Gaudi) / v2.4.1 (CPU). However there are several files where much older versions are used. GenAIExamples, old CPU/rocm versions: ```...

Battlemage perf in Phoronix benchmarks

5

Phoronix compared B580 performance to other GPUs using Git version of Linux 6.13 kernel, and compute-runtime v24.45.31740.9 + IGC v2.1.12. While B580 perf was already better than A580 perf in...

Users want some tool to monitor their GPUs. Currently there are no good options: * [XPUM](https://github.com/intel/xpumanager) supports officially (is tested with) only data center GPUs, and as result, its: *...