Splunk Technology Add-on
TA-amdgpu
ROCm GPU telemetry
Free for personal use ยท commercial license required
GPU
rocm-smi (or amdgpu sysfs for the zero-dep core)
Why TA-amdgpu exists
AMD GPUs deserve first-class Splunk coverage too. TA-amdgpu collects temperature, power, VRAM, ECC errors and XGMI link state, giving one GPU dashboard whether the fleet is red or green.
What it collects
- GPU temperature and power
- VRAM usage and ECC errors
- XGMI link state
Runs on stock tools
No daemons, no agents, no exotic collectors - it shells out to well-known CLIs or reads sysfs, and parses the result into Splunk events.
External tools used: rocm-smi (or amdgpu sysfs for the zero-dep core).