shyan_andras asked . 2021-10-20

Consecutive call of parsim constantly increases memory usage (Ubuntu)

Hi,
 
I am trying to perform a parameter sweep on my simulink model using parsim. However, after parsim has been called for several times (after ~24h) Matlab gets killed (I am using Ubuntu 20.04). The ubuntu terminal output from where I started Matlab just shows "Killed".
 
Taking a closer look I noticed that memory usage constantly increases. I stopped the program and shut down the parallel pool (delete(gcp('nocreate'))) which released some memory. I cleared the workspace and nothing is running anymore, however Matlab still takes 40% of the memory  and until now I could not find out where this comes from.
 
I tried to abstract the important commands to clarify the structure of my program and my usage of the commands (the original script has much more parameters etc). In short, in nested for-loops combining my parameter values I create an array of Simulink.SimulationInput objects containing batch_size objects, run the simulations using parsim, create another array, run parsim again and so on:
 
start_vals = [0.1, 0.08, 0.06, 0.04, 0.03, 0.02];
stop_vals = [0.04, 0.03, 0.02, 0.01, 0.0, -0.05];
batch_size = 50;
model = 'modelName';

cnt_exp = 0;
cnt_file = 0;
cnt_batch = 0;

for start_val = start_vals
    for stop_val = stop_vals
        cnt_exp = cnt_exp +1;
        
        MP = setParam(MP,start_val, stop_val)
        simulation_length = 12;
        cnt_batch = cnt_batch +1;
        sim_input(cnt_batch) = Simulink.SimulationInput(model);
        sim_input(cnt_batch) = sim_input(cnt_batch).setVariable('simulation_length',simulation_length);
        sim_input(cnt_batch) = sim_input(cnt_batch).setVariable('MP',MP);
        
        if cnt_batch < batch_size
            total_runs = cnt_file * batch_size + cnt_exp;
            if total_runs ~= sum_exp % last incomplete batch has to be processed
                continue;
            end
        end
        
        total_runs = cnt_file * batch_size + cnt_exp;
        try
            simOut = parsim(sim_input);
            for res = 1:length(simOut)
                % Process result
            end
            % Save results
            cnt_file = cnt_file +1;
            
            % Reset counters and arrays
            cnt_exp = 0;
            cnt_batch = 0;
            clear sim_input;
            
        catch error
            warning ("Something went wrong")
        end
    end
end

I'd greatly appreciate any help or hint on what is going wrong here!

parsim , linux , memory , simulink , parallel computing... , parfor

Expert Answer

John Williams answered . 2024-04-22 11:58:21

I managed to solve the issue by adapting the usage of sim_input and simOut. As clear apparently does not free the memory in Unix systems and delete somehow didn't work with these arrays, I avoided using clear and I initialized sim_input outside of the for-loop. Furthermore, I put all calculations concerning simOut into a function, hoping that this additional scope would clear the memory. I didn't check the effect of both measures separately, but I think mainly sim_input was causing the problem.
 
So this is approximately how the code without memory leak looks like now:
 
 
start_vals = [0.1, 0.08, 0.06, 0.04, 0.03, 0.02];
stop_vals = [0.04, 0.03, 0.02, 0.01, 0.0, -0.05];
batch_size = 50;
model = 'modelName';

cnt_exp = 0;
cnt_file = 0;
cnt_batch = 0;

for i=1:batch_size
    sim_input(i) = Simulink.SimulationInput(model);
end

for start_val = start_vals
    for stop_val = stop_vals
        cnt_exp = cnt_exp +1;
        
        MP = setParam(MP,start_val, stop_val)
        simulation_length = 12;
        cnt_batch = cnt_batch +1;
        sim_input(cnt_batch) = sim_input(cnt_batch).setVariable('simulation_length',simulation_length);
        sim_input(cnt_batch) = sim_input(cnt_batch).setVariable('MP',MP);
        
        if cnt_batch < batch_size
            total_runs = cnt_file * batch_size + cnt_exp;
            if total_runs ~= sum_exp % last incomplete batch has to be processed
                continue;
            end
        end
        
        total_runs = cnt_file * batch_size + cnt_exp;
        try
            result_arr = callParSim(sim_input);
            % Save results
            cnt_file = cnt_file +1;
            
            % Reset counters and arrays
            cnt_exp = 0;
            cnt_batch = 0;
            
        catch error
            warning ("Something went wrong")
        end
    end
end

function result_arr = callParSim(sim_input)
simOut = parsim(sim_input);

for res = 1:length(simOut)
    %Process result, calculate result_arr
end

end


Not satisfied with the answer ?? ASK NOW

Frequently Asked Questions

MATLAB offers tools for real-time AI applications, including Simulink for modeling and simulation. It can be used for developing algorithms and control systems for autonomous vehicles, robots, and other real-time AI systems.

MATLAB Online™ provides access to MATLAB® from your web browser. With MATLAB Online, your files are stored on MATLAB Drive™ and are available wherever you go. MATLAB Drive Connector synchronizes your files between your computers and MATLAB Online, providing offline access and eliminating the need to manually upload or download files. You can also run your files from the convenience of your smartphone or tablet by connecting to MathWorks® Cloud through the MATLAB Mobile™ app.

Yes, MATLAB provides tools and frameworks for deep learning, including the Deep Learning Toolbox. You can use MATLAB for tasks like building and training neural networks, image classification, and natural language processing.

MATLAB and Python are both popular choices for AI development. MATLAB is known for its ease of use in mathematical computations and its extensive toolbox for AI and machine learning. Python, on the other hand, has a vast ecosystem of libraries like TensorFlow and PyTorch. The choice depends on your preferences and project requirements.

You can find support, discussion forums, and a community of MATLAB users on the MATLAB website, Matlansolutions forums, and other AI-related online communities. Remember that MATLAB's capabilities in AI and machine learning continue to evolve, so staying updated with the latest features and resources is essential for effective AI development using MATLAB.

Without any hesitation the answer to this question is NO. The service we offer is 100% legal, legitimate and won't make you a cheater. Read and discover exactly what an essay writing service is and how when used correctly, is a valuable teaching aid and no more akin to cheating than a tutor's 'model essay' or the many published essay guides available from your local book shop. You should use the work as a reference and should not hand over the exact copy of it.

Matlabsolutions.com provides guaranteed satisfaction with a commitment to complete the work within time. Combined with our meticulous work ethics and extensive domain experience, We are the ideal partner for all your homework/assignment needs. We pledge to provide 24*7 support to dissolve all your academic doubts. We are composed of 300+ esteemed Matlab and other experts who have been empanelled after extensive research and quality check.

Matlabsolutions.com provides undivided attention to each Matlab assignment order with a methodical approach to solution. Our network span is not restricted to US, UK and Australia rather extends to countries like Singapore, Canada and UAE. Our Matlab assignment help services include Image Processing Assignments, Electrical Engineering Assignments, Matlab homework help, Matlab Research Paper help, Matlab Simulink help. Get your work done at the best price in industry.