Skip to content

Updating readme #800

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

Closed
wants to merge 2 commits into from
Closed

Updating readme #800

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
80d42d8
Updating readme
anupambhatnagar Aug 22, 2023
6b34912
fix typos
anupambhatnagar Aug 22, 2023
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
23 changes: 18 additions & 5 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,20 +2,33 @@

Kineto is part of the PyTorch Profiler.

The Kineto project was started to help enable
The Kineto project enables:
- **performance observability and diagnostics** across common ML bottleneck components
- **actionable recommendations** for common issues
- integration of external system-level profiling tools
- integration with popular visualization platforms and analysis pipelines

A central component is libkineto, a profiling library with special focus on low-overhead GPU timeline tracing.

The PyTorch Profiler TensorBoard plugin provides powerful and intuitive visualizations of profiling results, as well as actionable recommendations, and is the best way to experience the new PyTorch Profiler.
A central component is Libkineto, a profiling library with special focus on low-overhead GPU timeline tracing.

## Libkineto

Libkineto is an in-process profiling library integrated with the PyTorch Profiler. Please refer to the [README](libkineto/README.md) file in the `libkineto` folder as well as documentation on the [new PyTorch Profiler API](https://pytorch.org/docs/master/profiler.html).

## PyTorch TensorBoard Profiler
## Holistic Trace Analysis

Holistic Trace Analysis (HTA) is an open source performance debugging library aimed at
distributed workloads. HTA takes as input PyTorch Profiler traces and elevates the performance
bottlenecks to enable faster debugging. Here's a partial list of features in HTA:

1. [Temporal Breakdown](https://hta.readthedocs.io/en/latest/source/features/temporal_breakdown.html): Breakdown of GPU time in terms of time spent in computation, communication, memory events, and idle time on a single node and across all ranks.
1. [Idle Time Breakdown](https://hta.readthedocs.io/en/latest/source/features/idle_time_breakdown.html): Breakdown of GPU idle time into waiting for the host, waiting for another kernel or attributed to an unknown cause.
1. [Kernel Breakdown](https://hta.readthedocs.io/en/latest/source/features/kernel_breakdown.html): Find kernels with the longest duration on each rank.
1. [Kernel Duration Distribution](https://hta.readthedocs.io/en/latest/source/features/kernel_breakdown.html#kernel-duration-distribution): Distribution of average time taken by longest kernels across different ranks.
1. [Communication Computation Overlap](https://hta.readthedocs.io/en/latest/source/features/comm_comp_overlap.html): Calculate the percentage of time when communication overlaps computation.

For a complete list see [here](http://hta.readthedocs.io).

## PyTorch TensorBoard Profiler (Deprecated)
The goal of the PyTorch TensorBoard Profiler is to provide a seamless and intuitive end-to-end profiling experience, including straightforward collection from PyTorch and insightful visualizations and recommendations in the TensorBoard UI.
Please refer to the [README](tb_plugin/README.md) file in the `tb_plugin` folder.

Expand Down