Skip to content

Commit

Permalink
Fixes typo
Browse files Browse the repository at this point in the history
  • Loading branch information
erikhuck committed Jun 11, 2024
1 parent e0ae0e2 commit 4ec54d2
Show file tree
Hide file tree
Showing 2 changed files with 3 additions and 3 deletions.
4 changes: 2 additions & 2 deletions docs/notebook/tutorial.ipynb
Original file line number Diff line number Diff line change
Expand Up @@ -144,7 +144,7 @@
"\n",
"The CPU utilization includes the system core count field which is the total number of cores available system-wide. Utilization is measured for the main process, its descendents, the main process and its descendents combined, and CPU utilization across the entire system. The sum percent is the sum of the percentages of all the cores being used. The hardware percent is that divided by the expected number of cores being used i.e. the optional `n_expected_cores` parameter (defaults to the number of cores in the entire system) for the main, descendents, and combined measurements. For the system measurements, hardware percent is divided by the total number of cores in the system regardless of the value of `n_expected_cores`. The max percent is the highest percentage detected through the duration of tracking while the mean percent is the average of all the percentages detected over that duration. The CPU utilization concludes with the maximum number of threads used at any time for the main process and the sum of the threads used across its descendent processes and combined.\n",
"\n",
"The GPU utilization is similar to the CPU utilization but rather than being based on utilization of processes, it can only measure the utilization percentages of the GPUs themselves, regardless of what processes are using them. To ameliorate this limiation, the optional `gpu_uuids` parameter can be set to specify which GPUs to measure utilization for (defaults to all the GPUs in the system). The system GPU count is the total number of GPUs in the system. The sum percent is the sum of all the percentages of these GPUs and the hardware percent is that divided by the expected number of GPUs being used (i.e. `len(gpu_uuids)`). Likewise with CPU utilization, the max and mean of both the sum and hardware percentages are provided.\n",
"The GPU utilization is similar to the CPU utilization but rather than being based on utilization of processes, it can only measure the utilization percentages of the GPUs themselves, regardless of what processes are using them. To ameliorate this limitation, the optional `gpu_uuids` parameter can be set to specify which GPUs to measure utilization for (defaults to all the GPUs in the system). The system GPU count is the total number of GPUs in the system. The sum percent is the sum of all the percentages of these GPUs and the hardware percent is that divided by the expected number of GPUs being used (i.e. `len(gpu_uuids)`). Likewise with CPU utilization, the max and mean of both the sum and hardware percentages are provided.\n",
"\n",
"The compute time is the real time that the computational-resource tracking lasted (as compared to CPU time)."
]
Expand Down Expand Up @@ -1189,7 +1189,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.12"
"version": "3.12.2"
}
},
"nbformat": 4,
Expand Down
2 changes: 1 addition & 1 deletion docs/tutorial.rst
Original file line number Diff line number Diff line change
Expand Up @@ -148,7 +148,7 @@ processes and combined.
The GPU utilization is similar to the CPU utilization but rather than
being based on utilization of processes, it can only measure the
utilization percentages of the GPUs themselves, regardless of what
processes are using them. To ameliorate this limiation, the optional
processes are using them. To ameliorate this limitation, the optional
``gpu_uuids`` parameter can be set to specify which GPUs to measure
utilization for (defaults to all the GPUs in the system). The system GPU
count is the total number of GPUs in the system. The sum percent is the
Expand Down

0 comments on commit 4ec54d2

Please sign in to comment.