Improving performance of parallel code sections

1 次查看（过去 30 天）

A.B. 2024-1-31

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2076861-improving-performance-of-parallel-code-sections

编辑： Matt J 2024-2-1

Consider the following test code which benchmarks different methods of parallelization within MATLAB:

function partest()
    
    delete(gcp("nocreate"));
    pool = parpool("threads", 3);
    njob = pool.NumWorkers;
    ndim = 10000000;
    state = rand(ndim, njob);
    dummy = 0;
    
    % parfor
    tic
    func = zeros(njob, 1);
    parfor ijob = 1 : njob
        func(ijob) = getFunc(state(:, ijob));
    end
    disp("  parfor = " + toc + " seconds.")
    dummy = dummy + sum(func);
    
    % spmd
    tic
    func = zeros(njob, 1);
    spmd
        indx = spmdIndex;
        func(indx) = getFunc(state(:, indx));
    end
    disp("    spmd = " + toc + " seconds.")
    %dummy = dummy + sum([func{:}]);
    
    % parfeval
    tic
    clear func
    fout(1 : njob) = parallel.FevalFuture;
    for ijob = 1 : njob
        fout(ijob) = parfeval(pool, @getFunc, 1, state(:, ijob));
    end
    func = fetchOutputs(fout);
    disp("parfeval = " + toc + " seconds.")
    dummy = dummy + sum(func);
    
    % serial
    tic
    func = zeros(njob, 1);
    for ijob = 1 : njob
        func(ijob) = getFunc(state(:, ijob));
    end
    disp("  serial = " + toc + " seconds.")
    dummy = dummy + sum(func);
    disp(dummy)
    delete(gcp("nocreate"));
end
function func = getFunc(state)
    %func = -sum(state.^2);
    func = 0;
    for idim = 1 : length(state)
        func = func - state(idim)^2;
    end
end

Here is the benchmark output:

  parfor = 0.092802 seconds.
    spmd = 0.044583 seconds.
parfeval = 0.032457 seconds.
  serial = 0.050159 seconds.

Firstly, why is parfeval outperforming all others?

Secondly, are there anything that could be done to any of these parallel constructs to improve their performance against the last (serial) case?

2 个评论
显示无隐藏无

Edric Ellis 2024-2-1

Can I ask

What OS are you using?
Which release of MATLAB?
How many physical cores do you have - what's the result of maxNumCompThreads.

A.B. 2024-2-1

WSL, 2023b, 12 per cpu, 12

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Matt J 2024-1-31

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2076861-improving-performance-of-parallel-code-sections#answer_1400831

编辑：Matt J 2024-1-31

The task is too small for parallelization to have any meaningful effect. Similarly, the relative performance numbers for the different methods is not meaningful. You need a task that is at least 30 seconds long serially for any of the comparisons to make any sense.

9 个评论
显示 7更早的评论隐藏 7更早的评论

A.B. 2024-2-1

No, virtually. That was the response to the strangeness issue I mentioned and you disregaded as normal. The real issue which prompted this post is the performance difference between the constructs, which of course diminishes with workload size. I just moved on with what's pragmatic. I'd appreciate your explanation of the phenomenon if you have any.

Matt J 2024-2-1

编辑：Matt J 2024-2-1

What about my earlier comment about having more than 1 loop iteration per worker? It makes intuitive sense to me that parFeval would be optimal when you have only 1 function call per worker. Otherwise, if parfor were optimal for everything, why would they provide parFeval?

请先登录，再进行评论。

请先登录，再回答此问题。

类别

Parallel Computing Parallel Computing Toolbox Parallel Computing Fundamentals

在 Help Center 和 File Exchange 中查找有关 Parallel Computing Fundamentals 的更多信息

产品

版本

R2023b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by