I’m trying to speed up some code that takes subsets of columns from a data matrix and calculates a maximum score. The orginal code is complicated, so I’ve tried to simplfy it was much as possible here: using Combinatorics #Helper function to loop through pairs of data columns allpairs(v) = Iterato…

It would presumably help to squeeze out all the allocations from your score function. I’ve been looking for an excuse to play with @Mason ’s neat Bumper.jl as an alternative to writing technical multithreaded code that manually manages local buffers correctly, and this seems like a good opportunity. …

Having issues speeding up code with multithreading

General Usage Performance

LaurentPlagne July 12, 2023, 3:44pm 10

And all the time is spent in allocation required by the linear solve:

2 Likes

Topic		Replies	Views
Speeding up maximization Performance memory-allocation	4	674	January 24, 2022
Scaling of @threads for "embarrassingly parallel" problem Performance threads	29	1958	January 20, 2023
Data structures for threaded computing Performance	23	2844	October 23, 2019
Poor performance multiplying many (large) matrices multithreaded Performance question , linearalgebra	11	2477	July 13, 2020
Julia multithreading is running slower than serial, can someone please explain why…? Performance multithreading , floops	14	705	March 24, 2023

Having issues speeding up code with multithreading

Related topics