DLProf Viewer User Guide

Abstract

DLProf Viewer User Guide provides instructions on how to use the DLProf Viewer to analyze performance results gathered by the NVIDIA Deep Learning Profiler.

1.1. Overview

The DLProf Viewer makes it easy to visualize the performance of your models by showing Top 10 operations that took the most time, eligibility of Tensor Core operations and Tensor Core usage, as well as interactive iteration reports.

1.2. What's New in 1.7.0

• New Group Ops view.
• Iteration Bar Charts can now be zoomed and vertically panned.

1.3. Features

This release includes these commands and features:

• Panelized Dashboard Summary View: A summary view comprising several panels that provide a quick overview of the performance results.
• Top-level Key Metrics: The Summary view displays several key metrics that are used to quickly gauge the quality of the performance, including Average Iteration Time and Tensor Core Utilization.
• Top 10 GPU Ops Node: A table in the Summary view lists the top 10 Op Nodes with the most time spent on the GPU.
• Expert Systems Panel: This panel displays any issues detected by the DLProf Expert Systems, along with suggestions on how to address the issues and improve the models performance.
• Iteration Summary Panel: This panel visually displays iterations. Users can quickly see how many iterations are in the model, the iterations that were aggregated/profiled, and the durations of tensor core kernels in each iteration.
• Interactive Tables: All tables in detailed views are completely interactive, allowing the use to sort, filter, and paginate the display.
• Interoperable Tables: Several views have the ability to drill down for more information. Selecting a row in one table will populate the next table with performance information pertaining to the selection.
• Client/server architecture:
  • All of the data is now in a relational database.
  • Load times have improved for all views.
• GPUs View: Displays GPU utilization and GPU properties of all GPUs used during profiling.

2.1. Installing Using Python Wheel

The DLPROF Viewer can be easily installed from the NVIDIA PY index.

Install nvidia-pyindex.

            

            
$ pip install nvidia-pyindex
        

Next install the DLProf Viewer using pip:

            

            
$ pip install nvidia-dlprofviewer
        

2.2. Using the NGC Docker Container

The following steps are required to use pre-built NGC containers:

• Ensure you have access and are logged into NGC. For step-by-step instructions, see the NGC Getting Started Guide.
• Install Docker and nvidia-docker. For DGX users, see Preparing to use NVIDIA Containers. For users other than DGX, see nvidia-docker installation documentation.

Use docker pull to get the TensorFlow container from NGC:

            

            
$ docker pull nvcr.io/nvidia/tensorflow:xx.yy>-tf1-py3
        

Assuming the training data for the model is available in /full/path/to/training/data, you can launch the container with the following command:

            

            
$ docker run --rm --gpus=1 --shm-size=1g --ulimit memlock=-1 \
--ulimit stack=67108864 -it -p8000:8000 -v/full/path/to/training/data:/data \
nvcr.io/nvidia/tensorflow:<xx.yy>-tf1-py3
        

2.3. Generating DLProf Database

The DLProf database is created directly from the Deep Learning Profiler. See the Deep Learning Profiler User Guide for information on how to generate the database.

2.4. Starting DLProf Viewer

The DLProf Viewer is installed in all containers that DLProf is installed in the NVIDIA GPU Cloud (NGC). The container must be run with the -p8000:8000 option to open port 8000 for the DLProf Viewer. Use any port such as 8001, 8002, etc.

Usage:

            

            
Usage: dlprofviewer [-h] [--version] [-b ADDRESS] [-p PORT] database
 
positional arguments:
  database              Specify path to DLProf database.
 
optional arguments:
  -h, --help            Show this help message and exit
  --version             Show program's version number and exit
  -b ADDRESS, --bind ADDRESS
                        Specify alternate bind address. Use '0.0.0.0' to serve to the entire local network. [default: localhost]
  -p PORT, --port PORT  Specify alternate port. [default: 8000]
        

The DLProf Viewer server is launched directly from the container:

            

            
$ dlprofviewer dlprof_dldb.sqlite
        

Where dlprof_dldb.sqlite is the path to the DLProf database.

2.5. Bind

If you want anyone on your local network to be able to access your dlprofviewer server in their browser, launch with -b 0.0.0.0:

            

            
$ dlprofviewer -b 0.0.0.0 dlprof_dldb.sqlite
        

Once the server is running, the DLProf app can be viewed in a browser using the URL:

            

            
http://<machine IP Address>:8000
        

2.6. Port

You can also specify a custom port with -p PORT, e.g. (port 8001):

            

            
$ dlprofviewer -b 0.0.0.0 -p 8001 dlprof_dldb.sqlite
        

Then use your browser to go to this URL:

            

            
http://<machine IP Address>:8000
        

2.7. How to Launch DLProf Viewer

The DLProf Viewer is installed in all containers that DLProf is installed in the NVIDIA GPU Cloud (NGC). The container must be run with the -p8000:8000 option to open port 8000 for the DLProf Viewer. Use any port such as 8001, 8002, etc.

2.8. Backwards Compatibility

Some older versions of the DLProf database can be used as input to the latest dlprofviewer:

• A DLProf database’s initial version is the version of DLProf used to create it.
• DLProf databases v1.3+ can be used as input to DLProf CLI v1.3+
  • The DLProf database version should still be less than or equal to the dlprofviewer version.
• DLProf databases v1.1 and v1.2 cannot be used as input to dlprofviewer v1.3+.
• When an older DLProf database is used as input, some minor changes will be made to it to allow it to be used correctly with the latest dlprofviewer.
• Backwards compatibility gets “reset” each major release.
  • dlprofviewer v2.x will not support DLProf databases v1.y.

This section describes each of the available views in the DLProf Viewer.

3.1. Terms and Definitions

3.2. Pane Overview

The DLProf Viewer user interface is divided into two panes:

Pane	Definintion
Content Pane	Otherwise known as the VIEWS pane, this is where you will see all of the different pieces of the profiled neural network.
Details Pane	This pane is to display additional details about the information selected in the content pane. Let’s say you’re looking at the Kernel Summary panel in the content pane, and you want to see more information about the kernels. Click on the button (tooltip = ‘Show Kernel Details’). The detailed kernel information will appear in the Details Pane. Click the ‘X’ button to remove the details pane.

3.3. Navigation Bar

This standard navigation bar is a navigation header that is placed at the top of the viewer. It contains a few buttons like most standard web sites. It does not scroll off the screen so these buttons are always available.

Some screen shots in this documentation might have a pulldown menu called Domain in the Navigation Bar. This feature has been removed from the DLProf Viewer.

UI Controls

3.4. Content Pane

3.4.1. Dashboard

The Dashboard view provides a high level summary of the performance results in a panelized view. This view serves as a starting point in analyzing the results and provides several key metrics.

3.4.1.1. GPU Utilization Panel

The GPU Idle panel visually indicates the percentage of GPU utilization time during execution of aggregated iterations. Hovering over a slice in the chart will show the numeric percentage.

Fields

UI Controls

3.4.1.2. Resource Usage Breakdown Panel

The panel provides a breakdown of profile activity into resource categories.

Fields

UI Controls

3.4.1.3. Total Kernel GPU Time Panel

The panel provides key metrics about the kernels in the network aggregated over the specific iteration range.

• Hovering over a slice in the chart will show the aggregated GPU time.
• Clicking a legend item will toggle its visualization in the chart.

Fields

UI Controls

Control	Definition
	Show Kernel Details panel in Details Pane.
Legend Label	Toggle between hiding and showing legend entry in chart.
More...	Show drop-down menu of more views.

3.4.1.4. Tensor Core Kernel Efficiency Panel

• Hovering over a slice in the chart will show the percentage.
• Clicking a legend item will toggle its visualization in the chart.

Performance Summary Panel

The Performance Summary panel provides top level key metrics about the performance data aggregated over the specific iteration range. A helpful tooltip text will appear when hovering over the ‘i’ icon.

3.4.1.6. Iteration Summary panel

This panel visually displays iterations. Users can quickly see how many iterations are in the model, the iterations that were aggregated/profiled, and the durations of tensor core kernels in each iteration. The colors on this panel match the colors on all the other dashboard panels.

For more information on this panel, see Iterations View.

3.4.1.7. Top 10 GPU Ops Panel

Top 10 GPU Ops table shows the top 10 operations with the largest execution times on the GPU. This table comes pre-sorted with the order of each row in descending GPU Time. The table is not sortable or searchable.

3.4.1.8. System Configuration Panel

3.4.1.9. Expert Systems panel

The Expert Systems panel displays common issues detected in the profiled network and provides potential solutions and suggestions to address the issues. The panel will only show issues that have been detected by DLProf. For a full list of potential issues that DLProf looks for, see the Expert Systems section in the Deep Learning Profiler User Guide.

Clicking on the double-green arrows shows additional information about the detected problem.

3.4.1.10. Guidance Panel

This panel provides static guidance to the user to help the user learn more about Tensor Cores, Mixed Precision training. The panel has hyperlinks for further reading. Clicking on a hyperlink inside the Guidance Panel will open a new tab in the browser.

3.4.2. Op Type Summary

This table aggregates metrics over all op types and enables users to see the performance of all the ops in terms of its types, such as Convolutions, Matrix Multiplications, etc.

See this description for all the features available in all DataTables.

Op Type Data Table

3.4.3. Group Ops

This view displays all operations in a hierarchical view. This new view is based on the ‘group node report’ provided by the DLProf CLI. Whereas the textual report provided by the CLI is a CSV file, this view is interactive by supporting filtering, sorting, and pagination. This information is interactive, it provides a powerful mechanism to quickly find ops and groups that consume too much duration.

See this description for all the features available in all DataTables.

The intersection of box black highlighting boxes above depicts the first eight of 32 ops in the level 17 group called ‘Conv2d::_conv_forward’. The values in the upper levels of the network graph are the sums of all the ops in the lower levels.

Definitions

Column	Definition
Group	An entity in the tree that represents a collection of ops.
Op	A normal operation in the network graph.
Level	Represents the depth of the group or op. Level one is the root. The higher the level value the deeper/lower the group or op is in the network graph.
or	Depict whether Group Op is Expanded or Collapsed. The indentation of the arrows depicts the level (or depth) of the group. Note: Note: There is a bug in the 3rd party treetable control whereby a group level 7 or higher appears in column one.
Group or Op	One of two values: Group or Op.
Name	The name of the selected row. A group has an appending slash. The indentation depicts the level (or depth) of the name in the network graph.
Level	Represents the depth of the group or op.
All other columns	If an op, the values represent counts and durations for that individual op . If a group, the values represent a summation of all the encapsulated ops

UI Controls

Selection Panel

Selection	Definition
Group
Op Name and Op ID match	(Not pictured: same as ‘Group’ above except the label is ‘Op Name Selected’)
Op Name and Op ID do not match

3.4.4. Ops and Kernels

This view enables users to view, search, sort all ops and their corresponding kernels in the entire network.

3.4.4.1. Ops Data Table

When a row is selected in the Ops table, a summary of each kernel for that op is displayed in the bottom table.

See this description for all the features available in all DataTables.

3.4.4.2. Kernel Summaries Data Table

See this description for all the features available in all DataTables.

Kernels Summary table

3.4.5. Kernels by Iteration

The Kernels by Iterations view shows operations and their execution time for every iteration. At a glance, you can compare iterations with respect to time as well as Kernels executed.

3.4.5.1. Iteration Summary Data Table

To see the kernels for a specific iteration, click a row in the top table. The ‘Kernels in Selected Iteration’ table will be filled with the kernels from the selection iteration.

See this description for all the features available in all DataTables.

3.4.5.2. Kernels in Selected Iteration

See this description for all the features available in all DataTables.

3.4.6. Kernels by Op

The Kernels by Op view is a variation of the Kernels by Iterations view. It has the capability to filter the list of kernels by iterations and op.

See this description for all the features available in all DataTables.

3.4.6.1. Iteration Summary Data Table

Selecting an iteration in the Iteration Summary table will populate the Ops in Selected Iteration table with all the profile data for the ops from the selected iteration. Selecting an op in the Ops in Selected Iteration table will populate the Kernels in Selected Op table with the list of kernels and timing data executed for the selected Op and Iteration.

See this description for all the features available in all DataTables.

3.4.6.2. Ops in Selected Iteration Table

Ops in Selected Iteration

See this description for all the features available in all DataTables.

3.4.6.3. Kernels Selected Iteration / Op Combination Table

Kernels in Selected Iteration / Op combination

See this description for all the features available in all DataTables.

3.4.7. Iterations view

This view displays iterations visually. Users can quickly see how many iterations are in the model, the iterations that were aggregated/profiled, and the accumulated durations of tensor core kernels in each iteration.

Fields

UI Controls

Workflow

Sometimes the original aggregation parameters on the DLProf command-line specified an iteration start value, an iteration stop value, or even key node that yielded a suboptimal representation of the network’s profile (such as including warm-up nodes). This feature allows the user to change those values and re-aggregate. Here is an example:

1. Notice in the Iterations View screenshot above:
   1. Iteration eight contains no work, and
   2. Iteration nine through fifteen (among others) contain way too much work. The aggregation values throughout the viewer contain those ops and kernels and potentially skews the results.
2. You can change the iteration start and iteration stop values in a number of ways: typing and clicking the up and down spinners. The best way is to hover over the field and spin the mouse wheel.
3. Some platforms do not have a default key node for a neural network. Sometimes the predefined key node is suboptimal. By clicking on the ‘Select Key Node’ picker, a key node can be selected. This is a full-featured panel that allows for filtering, sorting, and pagination to find a Key Node. See the view called Ops and Kernels for more details.
   Note:

   Note: There are fewer columns here than in the view, but the usability is the same.

   
   

   

4. Once any of these three fields change, click the Reaggregate button. A confirmation dialog is displayed:

   

5. After confirmation, a message is sent to the back-end DLProf server to reaggregate the profile with these values. This reaggregation process could potentially take minutes, so the following panel appears above all panels:

   

   The viewer is fully functional while the re-aggregation is taking place. The “Underway!” panel will remain in view above all other views and panels until reaggregation is complete.

   Note:

   Note: Do not click the browser’s BACK or REFRESH button. If either are accidentally, the “Underway!” panel will no longer appear. The re-aggregation will continue regardless.

6. When the reaggregation is complete, a “Success!” panel will appear like this:

   

7. At this point, clicking the REFRESH button will load all the newly reaggregated data into the viewer. If it’s not already there, the dropdown called Aggregation is displayed in the Navigation Bar:

   

3.4.8. GPUs View

This view shows the utilization of all GPUs during the profile run. It is broken down into two different but related elements:

• Bar Chart - Quick visualization where you can see the GPU utilization for every GPU used during the profile. This view appears only when more than one GPU is used in a profile.
• Table - This table shows a little more detail about each GPU device, including its name, Compute Capability, and SM count.

3.4.9. Details Pane

This pane shows more details about a particular area of the system.

3.4.9.1. Kernel Details Panel

This panel provides key metrics about the kernels in the network aggregated over the specific iteration range.

Fields

UI Controls

3.5. Using Data Tables

DataTables are used in many views in the DLProf Viewer. The features in DataTables enable users to quickly find information. Below is a screenshot to see the location of the features, followed by a table describing the functionality of each feature.

The database does not contain a legal configuration.

The DLProf Database cannot be found.

See Starting DLProf Viewer for more information.

An older database was found.

A new database was found.

Bad parameters found in URL.

Notice

This document is provided for information purposes only and shall not be regarded as a warranty of a certain functionality, condition, or quality of a product. NVIDIA Corporation (“NVIDIA”) makes no representations or warranties, expressed or implied, as to the accuracy or completeness of the information contained in this document and assumes no responsibility for any errors contained herein. NVIDIA shall have no liability for the consequences or use of such information or for any infringement of patents or other rights of third parties that may result from its use. This document is not a commitment to develop, release, or deliver any Material (defined below), code, or functionality.

NVIDIA reserves the right to make corrections, modifications, enhancements, improvements, and any other changes to this document, at any time without notice.

Customer should obtain the latest relevant information before placing orders and should verify that such information is current and complete.

NVIDIA products are sold subject to the NVIDIA standard terms and conditions of sale supplied at the time of order acknowledgement, unless otherwise agreed in an individual sales agreement signed by authorized representatives of NVIDIA and customer (“Terms of Sale”). NVIDIA hereby expressly objects to applying any customer general terms and conditions with regards to the purchase of the NVIDIA product referenced in this document. No contractual obligations are formed either directly or indirectly by this document.

NVIDIA products are not designed, authorized, or warranted to be suitable for use in medical, military, aircraft, space, or life support equipment, nor in applications where failure or malfunction of the NVIDIA product can reasonably be expected to result in personal injury, death, or property or environmental damage. NVIDIA accepts no liability for inclusion and/or use of NVIDIA products in such equipment or applications and therefore such inclusion and/or use is at customer’s own risk.

NVIDIA makes no representation or warranty that products based on this document will be suitable for any specified use. Testing of all parameters of each product is not necessarily performed by NVIDIA. It is customer’s sole responsibility to evaluate and determine the applicability of any information contained in this document, ensure the product is suitable and fit for the application planned by customer, and perform the necessary testing for the application in order to avoid a default of the application or the product. Weaknesses in customer’s product designs may affect the quality and reliability of the NVIDIA product and may result in additional or different conditions and/or requirements beyond those contained in this document. NVIDIA accepts no liability related to any default, damage, costs, or problem which may be based on or attributable to: (i) the use of the NVIDIA product in any manner that is contrary to this document or (ii) customer product designs.

No license, either expressed or implied, is granted under any NVIDIA patent right, copyright, or other NVIDIA intellectual property right under this document. Information published by NVIDIA regarding third-party products or services does not constitute a license from NVIDIA to use such products or services or a warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property rights of the third party, or a license from NVIDIA under the patents or other intellectual property rights of NVIDIA.

Reproduction of information in this document is permissible only if approved in advance by NVIDIA in writing, reproduced without alteration and in full compliance with all applicable export laws and regulations, and accompanied by all associated conditions, limitations, and notices.

THIS DOCUMENT AND ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY, “MATERIALS”) ARE BEING PROVIDED “AS IS.” NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE. TO THE EXTENT NOT PROHIBITED BY LAW, IN NO EVENT WILL NVIDIA BE LIABLE FOR ANY DAMAGES, INCLUDING WITHOUT LIMITATION ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL, PUNITIVE, OR CONSEQUENTIAL DAMAGES, HOWEVER CAUSED AND REGARDLESS OF THE THEORY OF LIABILITY, ARISING OUT OF ANY USE OF THIS DOCUMENT, EVEN IF NVIDIA HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Notwithstanding any damages that customer might incur for any reason whatsoever, NVIDIA’s aggregate and cumulative liability towards customer for the products described herein shall be limited in accordance with the Terms of Sale for the product.

HDMI

HDMI, the HDMI logo, and High-Definition Multimedia Interface are trademarks or registered trademarks of HDMI Licensing LLC.

OpenCL

OpenCL is a trademark of Apple Inc. used under license to the Khronos Group Inc.

Trademarks

NVIDIA, the NVIDIA logo, and cuBLAS, CUDA, cuDNN, DALI, DIGITS, DGX, DGX-1, DGX-2, DGX Station, DLProf, Jetson, Kepler, Maxwell, NCCL, Nsight Compute, Nsight Systems, NvCaffe, NVIDIA Ampere GPU Architecture, PerfWorks, Pascal, SDK Manager, Tegra, TensorRT, Triton Inference Server, Tesla, TF-TRT, and Volta are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. Other company and product names may be trademarks of the respective companies with which they are associated.