MATLAB system requirement for a faster Simulink output
5 vues (au cours des 30 derniers jours)
Afficher commentaires plus anciens
Nghi
le 20 Déc 2023
Réponse apportée : Sulaymon Eshkabilov
le 20 Déc 2023
Hello, my pc is 2 years old, gtx 1650, 16gb 2666mhz, r3 3100. I am running simulations on Simulink with numbers up to e10 or potentially more in the future. It takes some time for the program to give me a step response, especially I usually run for at least 200 seconds. For example, a pi controller that has gains up to 3e09, simulate for 200 seconds, takes about 30 seconds or more, for 1 run. Do you suggest me to upgrade my pc? If yes then which component should I improve? Cpu? My system is ryzen so only ryzen cpu. Thank you. I also wonder the effects of the hardware to the software operation, that's why I am asking
0 commentaires
Réponse acceptée
Sulaymon Eshkabilov
le 20 Déc 2023
To suggest you or somebody to upgare one's PC is quite subjective opinion I believe. Because the end user and who is paying for is not the person who is suggesting :).
Anyhow there are a few points to consider here to speed up the simulation process before investing in hardware components or upgrading your PC.
(1) Check what solver you are using.
(2) Check if your selected solver in your model is a fixed step or variable step size
(3) Check the possibility of employing Parallel Toolbox functions
(4) Check your MATLAB's memory management
(5) Check MATLAB's concurrency and if possible take the max number of available cores
(6) Check if your model uses any functions from external libraries
(7) Upgrade your MATLAB license to the latest version, e.g., MATLAB 2023b is much faster than MATLAB 2017a
(8) Use Profiliung to optimize your model to find out where your model spends most time.
If you still want to invest in your PC upgrade, then consider these points:
(1) Upgrade CPU. MATLAB simulations are bound to CPU. If possible, consider a multi-core CPU.
(2) Upgrade RAM. It helps faster and easier to handle datasets from your models.
(3) Upgrade GPU. MATLAB now supports GPU based computations.
(4) Upgrade or switch to SSD for data storage. SSD speeds up data writing and reading.
That is all.
0 commentaires
Plus de réponses (0)
Voir également
Catégories
En savoir plus sur GPU Computing dans Help Center et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!