Energy

The energy metrics are only available with Linaro Forge Ultimate. All metrics are measured per node. If you are running your job on more than one node, Linaro MAP shows the minimum, mean, and maximum power consumption of the nodes.

Note

Energy metrics are not available when linking to the static Linaro Forge sampler library.

NVIDIA

GPU power usage: The cumulative power consumption of all GPUs on the node, as measured by the NVIDIA on-board sensor. This metric is available if the Accelerator metrics are present.

Note

NVIDIA GPU power measurement requires an NVIDIA GPU that supports power monitoring. This can be checked on the command-line with nvidia-smi -q -d power. If the reported power values are reported as “N/A”, power monitoring is not supported.

AMD

GPU power usage: The average GPU power consumption on each node, as measured by the rocm_smi utility. This metric is available if the Accelerator metrics are present.

CPU

CPU power usage: The cumulative power consumption of all CPUs on the node, as measured by the Intel on-board sensor (Intel RAPL).

Note

CPU power measurement requires an Intel CPU with RAPL support, for example Sandy Bridge or newer, and the intel_rapl powercap kernel module to be loaded.

Cray

Cray PM CPU power usage: The cumulative power consumption of all CPUs on the node, as measured by the Cray Power Monitoring metrics (Cray PM).

System power usage: The power consumption of the node as measured by the Cray metrics.

Note

Node power monitoring is implemented through the Cray HSS energy counters. The Cray HSS energy counters are known to be available on Cray XK6 and XC30 machines.