(fwd) Integral sorting
H. -J. Werner
werner at theochem.uni-stuttgart.de
Thu Sep 5 16:56:33 BST 2002
You hit an integer overflow (integral record > 16 GB)!
The problem should not occur any more in molpro2002.3
Joachim Werner
----- Forwarded message from Akyl Tulegenov <pcxast at nottingham.ac.uk> -----
From: Akyl Tulegenov <pcxast at nottingham.ac.uk>
To: molpro-user at stchem.bham.ac.uk
Subject: Integral sorting
Date: Thu, 5 Sep 2002 16:32:55 +0100 (BST)
Message-ID: <Pine.GSO.4.21.0209051628120.4269-100000 at granby.ccc.nottingham.ac.uk>
Dear All!
Is there any way to overcome the problem shown below without swithing to
direct algorithm? I am slightly confused by such a large amount of memory
requested for sorting.
....
Variable memory set to 80000000 words, buffer space 158000 words
....
Using spherical harmonics
....
Point group C1
....
NUCLEAR CHARGE: 32
NUMBER OF PRIMITIVE AOS: 586
NUMBER OF SYMMETRY AOS: 468
NUMBER OF CONTRACTIONS: 412 ( 412A )
NUMBER OF CORE ORBITALS: 4 ( 4A )
NUMBER OF VALENCE ORBITALS: 18 ( 18A )
....
Eigenvalues of metric
1 0.460E-05 0.103E-04 0.113E-04 0.189E-04 0.193E-04 0.309E-04 0.338E-04 0.390E-04
Time for two-electron integrals: 2671.57 SEC
957414224 INTEGRALS WRITTEN OUT IN ***** RECORDS ON RECORD 1290 OF FILE 1
NUMBER OF SORTED TWO-ELECTRON INTEGRALS:-675791715 BUFFER LENGTH: 12288
NUMBER OF SEGMENTS: 2 SEGMENT LENGTH:-675791715 RECORD LENGTH: 65536
Memory used in sort: 0.17 MW
SORT1 READ 957414224 AND WROTE 370726955 INTEGRALS IN***** RECORDS. CPU TIME: 653.05 SEC, REAL TIME: 865.40 SEC
insufficient memory available - require 675791715 have 79922076
the request was for real words
ERROR EXIT
Best regards,
Akyl Tulegenov
----- End forwarded message -----
--
Prof. Hans-Joachim Werner
Institute for Theoretical Chemistry
University of Stuttgart
Pfaffenwaldring 55
D-70569 Stuttgart, Germany
Tel.: (0049) 711 / 685 4400
Fax.: (0049) 711 / 685 4442
e-mail: werner at theochem.uni-stuttgart.de
More information about the Molpro-user
mailing list