Skip to content

CON-13656 expose gpu metrics #351

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
atelpis merged 2 commits into from
Mar 23, 2026
Merged

CON-13656 expose gpu metrics #351

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7a126d2
CON-13656 Expose NVIDIA GPU metrics
llDrLove Mar 12, 2026
9544293
CON-13656 Add AMD GPU metrics
llDrLove Mar 12, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
58 changes: 38 additions & 20 deletions cmd/do-agent/aggregation.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -129,6 +129,18 @@ var gpuAggregationSpec = map[string][]string{
"dcgm_fi_dev_power_violation": nvidiaAggregatedLabels,
"dcgm_fi_dev_thermal_violation": nvidiaAggregatedLabels,
"dcgm_fi_dev_ecc_dbe_agg_total": nvidiaAggregatedLabels,
"dcgm_fi_prof_sm_active": nvidiaAggregatedLabels,
"dcgm_fi_prof_pipe_fp16_active": nvidiaAggregatedLabels,
"dcgm_fi_prof_pipe_fp32_active": nvidiaAggregatedLabels,
"dcgm_fi_prof_pipe_fp64_active": nvidiaAggregatedLabels,
"dcgm_fi_dev_retired_sbe": nvidiaAggregatedLabels,
"dcgm_fi_dev_retired_dbe": nvidiaAggregatedLabels,
"dcgm_fi_dev_xid_errors": nvidiaAggregatedLabels,
"dcgm_fi_dev_ecc_sbe_vol_total": nvidiaAggregatedLabels,
"dcgm_fi_dev_ecc_dbe_vol_total": nvidiaAggregatedLabels,
"dcgm_fi_dev_ecc_sbe_agg_total": nvidiaAggregatedLabels,
"dcgm_fi_dev_pcie_replay_counter": nvidiaAggregatedLabels,
"dcgm_fi_dev_sm_clock": nvidiaAggregatedLabels,

// GPU Utilization metrics
"amd_gpu_prof_gui_util_percent": amdAggregatedLabels,
Expand Down Expand Up @@ -177,26 +189,32 @@ var gpuAggregationSpec = map[string][]string{
// PCIe bandwidth
"amd_pcie_bandwidth": amdAggregatedLabels,

"amd_gpu_ecc_uncorrect_total": amdAggregatedLabels,
"amd_pcie_replay_count": amdAggregatedLabels,
"amd_pcie_recovery_count": amdAggregatedLabels,
"amd_pcie_replay_rollover_count": amdAggregatedLabels,
"amd_pcie_max_speed": amdAggregatedLabels,
"amd_pcie_speed": amdAggregatedLabels,
"amd_gpu_prof_cpf_cpf_stat_stall": amdAggregatedLabels,
"amd_gpu_clock": amdAggregatedLabels,
"amd_gpu_violation_proc_hot_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_soc_thermal_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_ppt_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_hbm_thermal_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_vr_thermal_tracking_accumulated": amdAggregatedLabels,
"amd_gpu_violation_current_accumulated_counter": amdAggregatedLabels,
"amd_gpu_junction_temperature": amdAggregatedLabels,
"amd_gpu_power_usage": amdAggregatedLabels,
"amd_gpu_package_power": amdAggregatedLabels,
"amd_gpu_memory_temperature": amdAggregatedLabels,
"amd_gpu_gfx_activity": amdAggregatedLabels,
"amd_gpu_prof_sm_active": amdAggregatedLabels,
"amd_gpu_ecc_uncorrect_total": amdAggregatedLabels,
"amd_gpu_ecc_correct_total": amdAggregatedLabels,
"amd_pcie_replay_count": amdAggregatedLabels,
"amd_pcie_recovery_count": amdAggregatedLabels,
"amd_pcie_replay_rollover_count": amdAggregatedLabels,
"amd_pcie_max_speed": amdAggregatedLabels,
"amd_pcie_speed": amdAggregatedLabels,
"amd_pcie_nac_received_count": amdAggregatedLabels,
"amd_pcie_nack_sent_count": amdAggregatedLabels,
"amd_gpu_prof_cpf_cpf_stat_stall": amdAggregatedLabels,
"amd_gpu_clock": amdAggregatedLabels,
"amd_gpu_violation_proc_hot_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_processor_hot_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_soc_thermal_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_ppt_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_hbm_thermal_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_vr_thermal_tracking_accumulated": amdAggregatedLabels,
"amd_gpu_violation_socket_thermal_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_vr_thermal_residency_accumulated": amdAggregatedLabels,
"amd_gpu_violation_current_accumulated_counter": amdAggregatedLabels,
"amd_gpu_junction_temperature": amdAggregatedLabels,
"amd_gpu_power_usage": amdAggregatedLabels,
"amd_gpu_package_power": amdAggregatedLabels,
"amd_gpu_memory_temperature": amdAggregatedLabels,
"amd_gpu_gfx_activity": amdAggregatedLabels,
"amd_gpu_prof_sm_active": amdAggregatedLabels,
}

// DI metrics: drop high-cardinality labels we don't want to keep.
Expand Down
58 changes: 38 additions & 20 deletions cmd/do-agent/whitelist.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -180,6 +180,18 @@ var gpuWhitelist = map[string]bool{
"DCGM_FI_DEV_POWER_VIOLATION": true,
"DCGM_FI_DEV_THERMAL_VIOLATION": true,
"DCGM_FI_DEV_ECC_DBE_AGG_TOTAL": true,
"DCGM_FI_PROF_SM_ACTIVE": true,
"DCGM_FI_PROF_PIPE_FP16_ACTIVE": true,
"DCGM_FI_PROF_PIPE_FP32_ACTIVE": true,
"DCGM_FI_PROF_PIPE_FP64_ACTIVE": true,
"DCGM_FI_DEV_RETIRED_SBE": true,
"DCGM_FI_DEV_RETIRED_DBE": true,
"DCGM_FI_DEV_XID_ERRORS": true,
"DCGM_FI_DEV_ECC_SBE_VOL_TOTAL": true,
"DCGM_FI_DEV_ECC_DBE_VOL_TOTAL": true,
"DCGM_FI_DEV_ECC_SBE_AGG_TOTAL": true,
"DCGM_FI_DEV_PCIE_REPLAY_COUNTER": true,
"DCGM_FI_DEV_SM_CLOCK": true,

// GPU Utilization
"amd_gpu_prof_gui_util_percent": true,
Expand Down Expand Up @@ -228,26 +240,32 @@ var gpuWhitelist = map[string]bool{
// PCIe bandwidth
"amd_pcie_bandwidth": true,

"amd_gpu_ecc_uncorrect_total": true,
"amd_pcie_replay_count": true,
"amd_pcie_recovery_count": true,
"amd_pcie_replay_rollover_count": true,
"amd_pcie_max_speed": true,
"amd_pcie_speed": true,
"amd_gpu_prof_cpf_cpf_stat_stall": true,
"amd_gpu_clock": true,
"amd_gpu_violation_proc_hot_residency_accumulated": true,
"amd_gpu_violation_soc_thermal_residency_accumulated": true,
"amd_gpu_violation_current_accumulated_counter": true,
"amd_gpu_violation_ppt_residency_accumulated": true,
"amd_gpu_violation_hbm_thermal_residency_accumulated": true,
"amd_gpu_violation_vr_thermal_tracking_accumulated": true,
"amd_gpu_junction_temperature": true,
"amd_gpu_power_usage": true,
"amd_gpu_package_power": true,
"amd_gpu_memory_temperature": true,
"amd_gpu_gfx_activity": true,
"amd_gpu_prof_sm_active": true,
"amd_gpu_ecc_uncorrect_total": true,
"amd_gpu_ecc_correct_total": true,
"amd_pcie_replay_count": true,
"amd_pcie_recovery_count": true,
"amd_pcie_replay_rollover_count": true,
"amd_pcie_max_speed": true,
"amd_pcie_speed": true,
"amd_pcie_nac_received_count": true,
"amd_pcie_nack_sent_count": true,
"amd_gpu_prof_cpf_cpf_stat_stall": true,
"amd_gpu_clock": true,
"amd_gpu_violation_proc_hot_residency_accumulated": true,
"amd_gpu_violation_processor_hot_residency_accumulated": true,
"amd_gpu_violation_soc_thermal_residency_accumulated": true,
"amd_gpu_violation_current_accumulated_counter": true,
"amd_gpu_violation_ppt_residency_accumulated": true,
"amd_gpu_violation_hbm_thermal_residency_accumulated": true,
"amd_gpu_violation_vr_thermal_tracking_accumulated": true,
"amd_gpu_violation_socket_thermal_residency_accumulated": true,
"amd_gpu_violation_vr_thermal_residency_accumulated": true,
"amd_gpu_junction_temperature": true,
"amd_gpu_power_usage": true,
"amd_gpu_package_power": true,
"amd_gpu_memory_temperature": true,
"amd_gpu_gfx_activity": true,
"amd_gpu_prof_sm_active": true,
}
var diWhitelist = map[string]bool{
// DI GPU metrics
Expand Down
Loading