DOCA Documentation v2.7.0
1.0

NVIDIA DOCA DPA Statistics Tool

DOCA dpa-statistics is a CLI tool which allows users to monitor and obtain statistics on thread execution per running DPA process and thread. The tool is used to expose information about the running DPA processes and threads and to collect statistics on DPA thread performance.

The tool presents performance information for running DPA threads, including the number of cycles and instructions executed in a time period. The tool enables initiating and stopping collection of statistics and displaying the data collected per thread.

Info

The process ID output of the dpa-ps tool may be used as the input parameter for the dpa-statistics tool.

Info

This tool is supported for NVIDIA® BlueField®-3 only.

The command collect works on four mutually exclusive modes:

  • Enable mode – start collecting performance data

  • Disable mode – stop collecting performance data

  • Timeout mode – start collecting, wait with a timeout, stop collect and print info. User could break the wait with Ctrl-C command and then the timeout will be canceled and tool will disable statistics collection and prints the info with the actual time of the collect operation.

  • Infinite mode – no special flags. Same as timeout mode but with infinite timeout. The tool awaits the Ctrl-C command to stop.

The following table lists the collect command's flags and arguments:

Short Option

Long Option

Description

-h

--help

Help information

-d

--device

Device interface name (MST/RDMA)

-p

--process-id

Hexadecimal process ID for filtering

Info

This flag indicates a specific command for the command to operate on. Otherwise, statistics are collected from all processes.

-i

--suppress-header-info

Suppress print header info

-n

--enable

Enable collect info

-o

--disable

Disable collect info

-t

--timeout

Enable collect, wait with timeout, disable collect and print info

Info

Timeout value is in milliseconds.

Examples for inputting timeout value:

  • 45 – 45 milliseconds

  • 45.55 – 45 milliseconds and 550,000 nanoseconds

  • .0005 – 500 nanoseconds

  • 45m55n – 45 milliseconds and 55 nanoseconds

  • 66n – 66 nanoseconds

-r

--reset

Reset counters before operation starting collect operation

Presenting performance statistics is applicable after initiating data collection.

The following table lists the show command's flags and arguments:

Short Option

Long Option

Description

-h

--help

Help information

-d

--device

Device interface name (MST/RDMA)

-p

--process-id

Hexadecimal process ID for filtering

-i

--suppress-header-info

Suppress print header info

Output example:

Copy
Copied!
            

$ sudo ./dpa-statistics show -d mlx5_0 -p 1 ProcessID ThreadID Cycles Instruction Time Executions 1 3 266268 18193 164 41 4 411571 32727 252 47

Where:

  • ProcessID – The dpa_process_object_id to which the threads belongs

  • ThreadID – DPA thread object ID

  • Cycles – Total EU cycles the thread used

  • Instruction – Total number of instructions the thread executed

  • Time – Total time in ticks the thread was active

  • Executions – Total number of thread invocations

Examples

  • Example of collect in infinite mode for process 0 with suppress header info:

    Copy
    Copied!
                

    $ sudo ./dpa-statistics collect -d mlx5_0 -p 0 -i ...^C Data collected for 4606 milliseconds 0 nanoseconds 0 5 223964 13754 140 31 6 190130 13754 114 31

  • Example of collect in timeout mode with a timeout of 1 second and half a millisecond.

    Copy
    Copied!
                

    $ sudo ./dpa-statistics collect -d mlx5_0 -t 1000.500 Data collected for 1000 milliseconds 500000 nanoseconds ProcessID ThreadID Cycles Instruction Time Executions 0 5 223964 13754 140 31 6 190130 13754 114 31 1 3 266268 18193 164 41 4 411571 32727 252 47 2 3 0 223205 13754 137 31 1 189896 13754 113 31 2 191796 13754 117 31 4

  • Example of enabling statistics collection with reset of counters.

    Copy
    Copied!
                

    $ sudo ./dpa-statistics collect -d mlx5_0 -n -r

  • Example of disabling statistics collection.

    Copy
    Copied!
                

    $ sudo ./dpa-statistics collect -d mlx5_0 -o

  • Reading large statistics counter blocks takes a long time

  • The dpa-ps and dpa-statistics tools cannot be run at the same time on the same device

© Copyright 2024, NVIDIA. Last updated on Aug 15, 2024.