itanium crash with large molpro jobs
Ed Lee
E.P.Lee at soton.ac.uk
Fri Sep 19 14:09:48 BST 2003
Dear Molpro users,
We have just got molpro running on our itanium. I face crashes with large molpro jobs. The same type of job with a smaller basis set run alright, but larger jobs (same kind of job) with larger basis sets crash. We should have sufficient diskspace and memory on our itanium to run the job, as the same larger jobs run alright on other platforms available to us (SG and DEC). The last few entries in the output are as follows, but to an uninitiated like myself, these mean almost nothing. Has any one faced similar situation with itanium ? Any idea why it crashes ?
Best Regards, Ed. Lee
Contracted 2-electron integrals neglected if value below 1.0D-12
AO integral compression algorithm 1 Integral accuracy 1.0D-12
4262.724 MB (compressed) written to integral file ( 52.4%)
NUMBER OF SORTED TWO-ELECTRON INTEGRALS: 867393891. BUFFER LENGTH: 32768
NUMBER OF SEGMENTS: 28 SEGMENT LENGTH: 31997877 RECORD LENGTH: 524288
Memory used in sort: 32.55 MW
SORT1 READ 1016928851. AND WROTE 818617154. INTEGRALS IN 2355 RECORDS. CPU TIME: 433.91 SEC, REAL TIME: 1532.41 SEC
SORT2 READ 818617154. AND WROTE 867393891. INTEGRALS IN16261 RECORDS. CPU TIME: 234.38 SEC, REAL TIME: 880.26 SEC
FILE SIZES: FILE 1: 4268.8 MBYTE, FILE 4: 9877.6 MBYTE, TOTAL:14146.4 MBYTE
OPERATOR DM FOR CENTER 0 COORDINATES: 0.000000 0.000000 0.000000
ERROR EXIT
CURRENT STACK: MAIN
**********************************************************************************************************************************
DATASETS * FILE NREC LENGTH (MB) RECORD NAMES
1 19 3700.50 600 500 700 960 900 950 970 1000 1100 1400
BASINP VAR GEOM ABASIS SYMINP ZMAT AOBASIS BASIS S T
1410 1200 1210 1080 1600 129 1650 1300 1700
V H0 H01 AOSYM SMH P2S MOLCAS ERIS OPER
PROGRAMS * TOTAL RHF-SCF
CPU TIMES * 2315.92 2315.18
REAL TIME * 4661.47 SEC
DISK USED * 14.15 GB
**********************************************************************************************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20030919/67313026/attachment.html>
More information about the Molpro-user
mailing list