- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I am running code with RHS of X(148242,1) size, and total non zeros in the Sparse Matrix is A(11424868).
I am getting following error,
*** Error in PARDISO ( insufficient_memory) error_num= -200
*** Error in PARDISO memory allocation: FACT_ADR, size to allocate: 48 bytes
total memory wanted here: 201326595 kbyte
symbolic (max): 201326595 symbolic (permanent): 0
real(including 1 factor): 0
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
a.out 00000000004275A4 Unknown Unknown Unknown
a.out 0000000000416A77 Unknown Unknown Unknown
a.out 00000000004165E6 Unknown Unknown Unknown
a.out 0000000000415DC2 Unknown Unknown Unknown
a.out 00000000004155BC Unknown Unknown Unknown
Stack trace terminated abnormally.
I am running my code on cluster with node=4,ppn=4 and total memory 32gb.
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
mcmistry,
PARDISO works on SMP system only. There are no support Cluster version at this moment. Please try to launch your task on one node. Pay attention on the RAM size availble on your system and if the task size is not fit with this RAM, then use out-of-core mode (iparm(60)==2).
--Gennady
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thank you very much Gennady,
Is number of nonzeros in my sparse matrix is ok for Pardiso. That is 11424868, And If I put IPARM(60)=2, do I need to make any changes in the standard format of Pardiso code. I tried with what you suggested but it still give me the same error.
Mital
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Mital,
--Gennady
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks Gennady,
Will go through it..
Mital
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page