JoseSantosAMD 4424f320ad Kernel no overwrite (#303)
* Removing kernel name shortener from profile

-  Kernel name shortener now in analyze
-  db import uses pymongo to import

Signed-off-by: Jose Santos <josantos@amd.com>

* collections now added using pymongo

Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>

* kernel_name_shortener shortens dataframes and not a workload

-  argparser: --kernel-verbose to connection group
-  analysis_cli, profiler_rocprof_v2: remove kernel_name_shortener, so that it doesnt overwrite workload
-  db_import: demangle kernel_name dfs

Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>

* fix db name in pymongo command

Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>

* reformat

Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>

* fix typo

Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>

* remove checker for verbose in profiler_base

Signed-off-by: Jose Santos <josantos@amd.com>

* kernel_name_shortener now called in create_df_kerenel_top_stats and
create_df_pmc

Signed-off-by: Jose Santos <josantos@amd.com>

* add kernel_verbose to analysis_webui, remove stubbed values

Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>

* Update src/utils/kernel_name_shortener.py

keep console_debug

Co-authored-by: Cole Ramos <colramos@amd.com>
Signed-off-by: JoseSantosAMD <87447437+JoseSantosAMD@users.noreply.github.com>

* Update src/utils/kernel_name_shortener.py

Co-authored-by: Cole Ramos <colramos@amd.com>
Signed-off-by: JoseSantosAMD <87447437+JoseSantosAMD@users.noreply.github.com>

---------

Signed-off-by: Jose Santos <josantos@amd.com>
Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>
Signed-off-by: JoseSantosAMD <87447437+JoseSantosAMD@users.noreply.github.com>
Co-authored-by: Cole Ramos <colramos@amd.com>
2024-03-12 15:54:52 -05:00
2024-03-12 11:45:23 -05:00
2024-03-12 15:54:52 -05:00
2023-12-01 13:43:07 -06:00
2022-11-04 14:49:36 -05:00
2022-11-04 14:49:36 -05:00
2023-08-29 16:05:56 -05:00
2024-03-05 11:16:23 -06:00
2024-03-11 10:04:54 -05:00
2022-11-04 14:49:36 -05:00
2024-01-24 17:50:31 -06:00
2024-03-08 16:27:19 -06:00
2024-01-24 16:30:04 -06:00
2024-03-02 12:28:34 -06:00

Ubuntu 20.04 RHEL 8 MI100 Docs DOI

Omniperf

General

Omniperf is a system performance profiling tool for machine learning/HPC workloads running on AMD MI GPUs. The tool presently targets usage on MI100 and MI200 accelerators.

  • For more information on available features, installation steps, and workload profiling and analysis, please refer to the online documentation.

  • Omniperf is an AMD open source research project and is not supported as part of the ROCm software stack. We welcome contributions and feedback from the community. Please see the CONTRIBUTING.md file for additional details on our contribution process.

  • Licensing information can be found in the LICENSE file.

Development

Omniperf follows a main-dev branching model. As a result, our latest stable release is shipped from the main branch, while new features are developed in our dev branch.

Users may checkout dev to preview upcoming features.

How to Cite

This software can be cited using a Zenodo DOI reference. A BibTex style reference is provided below for convenience:

@software{xiaomin_lu_2022_7314631
  author       = {Xiaomin Lu and
                  Cole Ramos and
                  Fei Zheng and
                  Karl W. Schulz and
                  Jose Santos and
                  Keith Lowery and
                  Nicholas Curtis and
                  Cristian Di Pietrantonio},
  title        = {AMDResearch/omniperf: v1.1.0-PR1 (13 Oct 2023)},
  month        = oct,
  year         = 2023,
  publisher    = {Zenodo},
  version      = {v1.1.0-PR1},
  doi          = {10.5281/zenodo.7314631},
  url          = {https://doi.org/10.5281/zenodo.7314631}
}
S
Description
No description provided
Readme 282 MiB
Languages
C++ 67.5%
C 20.6%
Python 6.6%
CMake 3.4%
Shell 0.6%
Other 1.1%