parallel processing: job load- data too large too be saved

1 次查看（过去 30 天）

显示更早的评论

Isshaa Aarya 2017-5-14

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/340260-parallel-processing-job-load-data-too-large-too-be-saved

评论： Edric Ellis 2017-5-16

在 MATLAB Online 中打开

Hi Everyone,

I am using parallel processing in matlab.

I have looked at this tutorial https://au.mathworks.com/help/distcomp/introduction-to-parallel-solutions.html

and tried to repeat this for my algorithm.

My algorithm has multiple 3D and 4D matrices on which for loop is running to process the data.

I am trying to access this data with load(job,'myvariable')

But I keep getting this error.

Error using parallel.Job/load (line 33)
Error encountered while running the batch job. The error was:
The task result was too large to be stored.
Caused by:
  Error using parallel.internal.cluster.FileSerializer>iSaveMat (line 277)
  Data too large to be saved.

Can someone tell me what am I missing?

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Edric Ellis 2017-5-16

It appears as though your output arguments from your job are too large. You could try setting your default MAT-file version to -v7.3 as this supports larger data. See this section in the doc for instructions.

请先登录，再进行评论。

请先登录，再回答此问题。