Linaro Performance Reports is a simple tool used to generate a single-page HTML or plain text report that presents the overall performance characteristics of HPC applications. It supports pthreads, OpenMP, or MPI code on CPU, GPU, and MIC based architectures.

Availability and Restrictions

Versions

The versions currently available at OSC are:

You can use module spider linaro-pr to view available modules for a given machine. Feel free to contact OSC Help if you need other versions for your work.

Access

Linaro Performance Reports is available to all OSC users. We have 64 seats with 64 HPC tokens. Users can monitor the license status here.

Publisher/Vendor and License Type

Linaro, Commercial

Usage

Set-up

To load the module for the Linaro Performance Reports default version, use module load linaro-pr. To select a particular software version, use module load linaro-pr/version. For example, use module load linaro-pr/6.0 to load Linaro Performance Reports version 6.0, provided the version is available on the OSC cluster in use.

Using Linaro Performance Reports

You can use your regular executables to generate performance reports. The program can be used to analyze third-party code as well as code you develop yourself. Performance reports are normally generated in a batch job.

To generate a performance report for an MPI program:

module load linaro-pr
perf-report -np <num procs> --mpiargs="<mpi args>" <program> <program args>

where <num procs> is the number of MPI processes to use, <mpi args> represents arguments to be passed to mpiexec (other than -n or -np), <program> is the executable to be run and <program args> represents arguments passed to your program.

For example, if you normally run your program with mpiexec -n 12 wave_c, you would use

perf-report -np 12 wave_c

To generate a performance report for a non-MPI program:

module load linaro-pr
perf-report --no-mpi <program> <program args>

The performance report is created in both html and plain text formats. The file names are based on the executable name, number of processes, date and time, for example,  wave_c_12p_2016-02-05_12-46.html. To open the report in html format use

firefox wave_c_12p_2016-02-05_12-46.html

For more details, download the Linaro Performance Reports User Guide.

Performance Reports with GPU

Linaro Performance Reports can be used for CUDA codes. If you have an executable compiled with the CUDA library, you can launch Linaro Performance Reports with

perf-report {executable}

For more information, please read the section 6.10 of the Linaro Performance Reports User Guide.

Further Reading

• Linaro Performance Report homepage

See Also

Linaro MAP is a full scale profiler for HPC programs. We recommend using Linaro MAP after reviewing reports from Linaro Performance Reports. MAP supports pthreads, OpenMP, and MPI software on CPU, GPU, and MIC based architectures.

Availability & Restrictions

Versions

The Linaro MAP versions currently available at OSC are:

You can use module spider linaro-map to view available modules for a given machine. Feel free to contact OSC Help if you need other versions for your work.

Access

Linaro MAP is available to all OSC users. We have 64 seats with 80 HPC tokens. Users can monitor the Linaro License Server Status.

Publisher/Vendor and License Type

Linaro, Commercial

Usage

Set-up

To load the default version of the Linaro MAP module, use module load linaro-map. To select a particular software version, use module load linaro-map/version. For example, use module load linaro-map/6.0 to load Linaro MAP version 6.0, provided the version is available on the cluster in use. 

Note: Before you run MAP from the command line for the first time, open MAP as a GUI from OnDemand to configure with appropriate settings for your environment.

Using Linaro MAP

Profiling HPC software with Linaro MAP typically involves three steps: 

1. Prepare the executable for profiling.

Regular executables can be profiled with Linaro MAP, but source code line detail will not be available. You need executables with debugging information to view source code line detail: re-compile your code with a -g  option added among the other appropriate compiler options. For example:

mpicc wave.c -o wave -g -O3

This executable built with the debug flag can be used for Linaro Performance Reports as well.

Note: The -g flag turns off all optimizations by default. For profiling your code you should use the same optimizations as your regular executable, so explicitly include the -On flag, where n is your normal level of optimization, typically -O2 or -O3, as well as any other compiler optimization options.

2. Run your code to produce the profile data file (.map file).

Profiles are normally generated in a batch job.  To generate a MAP profile for an MPI program:

module load linaro-map
map --profile -np <num proc> --mpiargs="<mpi args>" <program> <program args>

where <num procs> is the number of MPI processes to use, <mpi args> represents arguments to be passed to srun (other than -n), <program> is the executable to be run and <program args> represents arguments passed to your program.

For example, if you normally run your program with mpiexec -n 12 wave_c, you would use

map --profile -np 12 wave_c

To profile a non-MPI program:

module load linaro-map
map --profile --no-mpi <program> <program args>

The profile data is saved in a .map file in your current directory.

As a result of this step, a .map file that is the profile data file is created in your current directory. The file name is based on the executable name, number of processes, date and time, for example, wave_c_12p_2016-02-05_12-46.map.

For more details on using Linaro MAP, refer to the Linaro Forge User Guide.

3. Analyze the profile data file using either the Linaro local client or the MAP GUI.

You can open the profile data file using a client running on your local desktop computer. For client installation and usage instructions, please refer to the section: Client Download and Setup. This option typically offers the best performance.

Alternatively, you can run MAP in interactive mode, which launches the graphical user interface (GUI).  For example:

map wave_c_12p_2016-02-05_12-46.map

For the GUI application, one should use an OnDemand VDI (Virtual Desktop Interface) or have X11 forwarding enabled (see Setting up X Windows). Note that X11 forwarding can be distractingly slow for interactive applications.

MAP with GPU

Linaro MAP can be used for CUDA codes. If you have an executable compiled with the CUDA library, you can launch Linaro MAP with

map {executable}

For more information, please read the Chapter 15 of the Linaro Forge User Guide.

Client Download and Setup

1. Download the client.

To download the client, go to the Linaro website and choose the appropriate Linaro Forge remote client download for Windows, Mac, or Linux. For Windows and Mac, just double click on the downloaded file and allow the installer to run. For Linux, extract the tar file using the command tar -xf file_name and run the installer in the extracted file directory with ./installer. Please contact OSC Help, if you have any issues on downloading the client.

2. Configure the client.

After installation, you can configure the client as follows:

• Open the client program. For Windows or Mac, just click the desktop icon or navigate to the application through its file path. For Linux use the command {linaro-forge-path}/bin/map.

• Once the program is launched, select Linaro MAP in the left column.
• In the Remote Launch drop down menu, select "Configure...".
• Click Add to create a new profile for your login.
• In the Host Name section, type your ssh connection. For example: "username@ruby.osc.edu".
• For Remote Installation Directory, type /usr/local/linaro/forge-{version}, specifying the Linaro Forge version number that created the data profile file you are attempting to view. For example, /usr/local/linaro/forge-7.0 for Linaro Forge version 7.0.
• You can test your login information by clicking Test Remote Launch. It will ask your password. Use the same password for the cluster login.
• Close the Configure window. You will see a new option under the Remote Launch drop down menu for the host name you entered. Select your profile and login with your password. 
• If the login was successful, then you should see License Serial:XXX in the bottom left corner of the window.

This login configuration is needed only for the first time of use. In subsequent times, you can just select your profile.

3. Open the profile data file.

After login, click on LOAD PROFILE DATA FILE. This opens a file browser of your home directory on the OSC cluster you logged onto. Go to the directory that contains the .map file and select it. This will open the file and allow you to navigate the source code line-by-line and investigate the performance characteristics. 

A license is not required to simply open the client, so it is possible to skip 2. Configure the client, if you download the profile data file to your desktop. You can then open it by just selecting LOAD PROFILE DATA FILE and navigating through a file browser on your local system.

Further Reading

• Linaro MAP homepage

See Also

Linaro DDT is a graphical debugger for HPC applications. It supports pthreads, OpenMP, or MPI code on CPU, GPU, and MIC based architectures.

Availability & Restrictions

Versions

The Linaro DDT versions currently available at OSC are:

You can use module spider linaro-ddt to view available modules for a given machine. Feel free to contact OSC Help if you need other versions for your work.

Access

Linaro DDT is available to all OSC users. We have 64 seats with 80 HPC tokens. Users can monitor the Linaro License Server Status.

Publisher/Vendor and License Type

Linaro, Commercial

Usage

Set-up

To load the module for the Linaro DDT default version, use module load linaro-ddt. To select a particular software version, use module load linaro-ddt/version. For example, use module load linaro-ddt/23.1 to load Linaro DDT version 23.1, provided the version is available on the OSC cluster in use.

Note: Before you run DDT from the command line for the first time, open DDT as a GUI from OnDemand to configure with appropriate settings for your environment.

Using Linaro DDT

DDT debugs executables to generate DDT reports. The program can be used to debug third-party code as well as code you develop yourself. DDT reports are normally generated in a batch job.

To generate a DDT report for an MPI program:

module load linaro-ddt
ddt --offline -np <num procs> --mpiargs="<mpi args>" <program> <program args>

where <num procs> is the number of MPI processes to use, <mpi args> represents arguments to be passed to mpiexec (other than -n or -np), <program> is the executable to be run and <program args> represents arguments passed to your program.

For example, if you normally run your program with mpiexec -n 12 wave_c, you would use

ddt --offline -np 12 wave_c

To debug a non-MPI program:

module load linaro-ddt
ddt --offline --no-mpi <program> <program args>

The DDT report is created in html format. The file names are based on the executable name, number of processes, date and time, for example, wave_c_12p_2016-02-05_12-46.html. To open the report use

firefox wave_c_12p_2016-02-05_12-46.html

Using the Linaro DDT GUI

To debug with the DDT GUI remove the --offline option. For example, to debug the MPI program above, use

ddt -np 12 wave_c

For a non-MPI program:

ddt --no-mpi <program> <program args>

This will open the DDT GUI, enabling interactive debugging options.

For the GUI application, one should use an OnDemand VDI (Virtual Desktop Interface) or have X11 forwarding enabled (see Setting up X Windows). Note that X11 forwarding can be distractingly slow for interactive applications.

For more details, see the Linaro DDT developer page.

DDT with GPU

DDT can be used for CUDA codes. If you have an executable compiled with the CUDA library, you can launch Linaro Performance Reports with

ddt {executable}

For more information, please read the chapter 14 of the Linaro Forge User Guide.