Skip to content

Commit

Permalink
Incorporate review comments
Browse files Browse the repository at this point in the history
Signed-off-by: Nicholas Curtis <nicurtis@amd.com>
  • Loading branch information
Nicholas Curtis authored and JoseSantosAMD committed Aug 15, 2023
1 parent 9cee849 commit f4bed5a
Show file tree
Hide file tree
Showing 6 changed files with 9 additions and 9 deletions.
2 changes: 1 addition & 1 deletion src/docs/analysis.md
Original file line number Diff line number Diff line change
Expand Up @@ -545,7 +545,7 @@ Connection Options:
**omniperf import for vcopy:**
```shell-session
$ omniperf database --import -H dummybox -u temp -t asw -w workloads/vcopy/mi200/
$ omniperf database --import -H pavii1 -u temp -t asw -w workloads/vcopy/mi200/
ROC Profiler: /usr/bin/rocprof
--------
Expand Down
4 changes: 3 additions & 1 deletion src/docs/conf.py
Original file line number Diff line number Diff line change
Expand Up @@ -54,7 +54,9 @@ def install(package):

myst_heading_anchors = 2
# enable replacement of (tm) & friends
myst_enable_extensions = ["replacements"]
myst_enable_extensions = [
"replacements"
]

# Add any paths that contain templates here, relative to this directory.
templates_path = ["_templates"]
Expand Down
2 changes: 1 addition & 1 deletion src/docs/high_level_design.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@

The [Omniperf](https://github.com/AMDResearch/omniperf) Tool is architecturally composed of three major components, as shown in the following figure.

- **Omniperf Profiling**: Acquire raw performance counters via application replay based on the [rocProfiler](https://rocm.docs.amd.com/projects/rocprofiler/en/latest/rocprof.html). The counters are stored in a comma-seperated value, for further analyis. A set of MI200 specific micro benchmarks are also run to acquire the hierarchical roofline data. The roofline model is not available on earlier accelerators.
- **Omniperf Profiling**: Acquire raw performance counters via application replay based on the [ROC Profiler](https://github.com/ROCm-Developer-Tools/rocprofiler). The counters are stored in a comma-seperated value, for further analyis. A set of MI200 specific micro benchmarks are also run to acquire the hierarchical roofline data. The roofline model is not available on earlier accelerators.

- **Omniperf Grafana Analyzer**:
- *Grafana database import*: All raw performance counters are imported into the backend MongoDB database for Grafana GUI analysis and visualization. Compatibility of previously generated data between Omniperf versions is not necessarily guarenteed.
Expand Down
4 changes: 2 additions & 2 deletions src/docs/installation.md
Original file line number Diff line number Diff line change
Expand Up @@ -172,9 +172,9 @@ $ sudo apt install ./mongodb-database-tools-ubuntu2004-x86_64-100.6.1.deb
```
> Installation instructions for alternative distributions can be found [here](https://www.mongodb.com/download-center/database-tools/releases/archive)
### 2) Persistent Storage
### Persistent Storage

The user will also bind MongoDB to a directory on the host OS to create a local backup in case of a crash or reset. In the Docker world, this is known as "creating a persistent volume":
The user may also choose to bind MongoDB to a directory on the host OS to create a local backup in case of a crash or reset:

```bash
$ sudo mkdir -p /usr/local/persist && cd /usr/local/persist/
Expand Down
2 changes: 1 addition & 1 deletion src/docs/introduction.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@

## Scope

MI Performance Profiler ([Omniperf](https://github.com/AMDResearch/omniperf)) is a system performance profiling tool for Machine Learning/HPC workloads running on AMD Instinct (tm) Accelerators. It is currently built on top of the [rocProfiler](https://rocm.docs.amd.com/projects/rocprofiler/en/latest/rocprof.html) to monitor hardware performance counters. The Omniperf tool primarily targets accelerators in the MI100 and MI200 families. Development is in progress to support MI300 and Radeon (tm) RDNA (tm) GPUs.
MI Performance Profiler ([Omniperf](https://github.com/AMDResearch/omniperf)) is a system performance profiling tool for Machine Learning/HPC workloads running on AMD Instinct (tm) Accelerators. It is currently built on top of the [ROC Profiler](https://github.com/ROCm-Developer-Tools/rocprofiler) to monitor hardware performance counters. The Omniperf tool primarily targets accelerators in the MI100 and MI200 families. Development is in progress to support MI300 and Radeon (tm) RDNA (tm) GPUs.

## Features

Expand Down
4 changes: 1 addition & 3 deletions src/docs/profiling.md
Original file line number Diff line number Diff line change
Expand Up @@ -328,9 +328,7 @@ ROCProfiler: input from "/tmp/rpl_data_230411_170300_29696/input0.xml"
```

#### Dispatch Filtering
Dispatch filtering is based on the *global* dispatch index of kernels in a run.

The following example profiles only the 0th dispatched kernel:
The following example demonstrates profiling on selected dispatches:
```shell-session
$ omniperf profile --name vcopy -d 0 -- ./vcopy 1048576 256
Resolving rocprof
Expand Down

0 comments on commit f4bed5a

Please sign in to comment.