why the caculation results is different for parfor-loop and for-loop?
조회 수: 2 (최근 30일)
이전 댓글 표시
load('matlab.mat')
for i =1:1
b_OLS(i) = X_*y;
end
parfor i =1:1
b_OLS_(i) = X_*y;
end
error = b_OLS_-b_OLS
The results is:
error =
1.8190e-11
why the caculation results is different for parfor-loop and for-loop?
댓글 수: 1
Stephen23
2021년 9월 27일
Because the associative laws of algebra do not generally hold for floating-point numbers.
답변 (1개)
Matt J
2021년 9월 23일
편집: Matt J
2021년 9월 27일
Probably because, with a parpool active, the matrix multiplication code cannot multithread the operation in precisely the same way. The vectors are split into parallel blocks of one size if a parpool is open and another size if not.
댓글 수: 1
Edric Ellis
2021년 9월 27일
In particular, by default for process-based parpool, the workers run in single-computational-thread mode. This can definitely result in slightly different results compared to multithreaded mode. You can use maxNumCompThreads(1) to put the client in single-computational-thread mode to check. Like this:
A = rand(1,10000);
B = rand(10000,1);
maxNumCompThreads(4);
c1 = A*B;
maxNumCompThreads(1);
c2 = A*B;
c1-c2
참고 항목
카테고리
Help Center 및 File Exchange에서 Parallel Computing Fundamentals에 대해 자세히 알아보기
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!