host.parquet

host.parquet has one row per host per sample. It describes the state, utilization and power draw of each host.

Column	Type	Unit	Description
timestamp	int64	ms	The time of the sample, since the start of the simulation.
timestamp_absolute	int64	ms	The time of the sample, since the Unix epoch.
host_name	string		The name of the host.
cluster_name	string		The name of the cluster of the host.
core_count	int32	count	The number of CPU cores of the host.
mem_capacity	int64	MiB	The memory of the host.
tasks_terminated	int32	count	The number of terminated tasks on the host.
tasks_running	int32	count	The number of tasks running on the host.
tasks_error	int32	count	The number of tasks on the host that failed.
tasks_invalid	int32	count	The number of tasks on the host in an unexpected state.
cpu_capacity	float	MHz	The total CPU capacity of the host.
cpu_usage	float	MHz	The CPU capacity used by the tasks.
cpu_demand	float	MHz	The CPU capacity the tasks ask for.
cpu_utilization	float	ratio	cpu_usage divided by cpu_capacity.
cpu_time_active	int64	ms	The time the CPU was active, since the previous sample.
cpu_time_idle	int64	ms	The time the CPU was idle, since the previous sample.
cpu_time_steal	int64	ms	The time tasks wanted to run but no CPU capacity was available, since the previous sample.
cpu_time_lost	int64	ms	The CPU time lost to interference, since the previous sample.
power_draw	float	W	The power draw of the host.
energy_usage	float	J	The energy the host used since the previous sample.
embodied_carbon	float	g	The embodied carbon of the host, for the time since the previous sample.
uptime	int64	ms	The time the host was up, since the previous sample.
downtime	int64	ms	The time the host was down, since the previous sample.
boot_time	int64	ms	When the host booted.

GPU columns

When the topology has GPUs, the file has a set of columns for each GPU of a host. The columns of the first GPU end in _0, those of the second in _1, and so on.

Column	Type	Unit	Description
gpu_capacity_i	float	MHz	The capacity of the GPU.
gpu_usage_i	float	MHz	The GPU capacity used by the tasks.
gpu_demand_i	float	MHz	The GPU capacity the tasks ask for.
gpu_utilization_i	float	ratio	gpu_usage_i divided by gpu_capacity_i.
gpu_time_active_i	int64	ms	The time the GPU was active, since the previous sample.
gpu_time_idle_i	int64	ms	The time the GPU was idle, since the previous sample.
gpu_time_steal_i	int64	ms	The time tasks wanted to run but no GPU capacity was available, since the previous sample.
gpu_time_lost_i	int64	ms	The GPU time lost to interference, since the previous sample.
gpu_power_draw_i	float	W	The power draw of the GPU.