Below are the results of running the OpenMP Stream on our 256 processor Altix.
hardware and software
----------------------------
> uname -a
Linux ram1 2.4.21-sgi230r3steiner #1 SMP Thu Sep 18 16:23:15 CDT 2003 ia64 unknown
256 1300 MHz Itanium 2 Rev. 5 Processor
Main memory size: ~1 Tbyte
> efc -V
Intel(R) Fortran Itanium(R) Compiler for Itanium(R)-based applications
Version 7.1, Build 20030814
Copyright (C) 1985-2003 Intel Corporation. All rights reserved.
GNU ld version 2.11.90.0.8 (with BFD 2.11.90.0.8)
Supported emulations:
elf64_ia64
> ecc -V
Intel(R) C++ Itanium(R) Compiler for Itanium(R)-based applications
Version 7.1, Build 20030814
Copyright (C) 1985-2003 Intel Corporation. All rights reserved.
GNU ld version 2.11.90.0.8 (with BFD 2.11.90.0.8)
Supported emulations:
elf64_ia64
Makefile flags
----------------
F77= efc
CC= ecc
OPTIMIZE = -O3
MP= -openmp
FFLAGS = -i8 -extend_source $(BASEFLAGS) $(OPTIMIZE) $(MP) $(OPTIONS)
CFLAGS = $(BASEFLAGS) $(OPTIMIZE) $(MP)
Output
--------
fm9@ram1> more best.256
fm9@ram1> export OMP_NUM_THREADS=254
fm9@ram1> dplace -x6 -c2-255 ./stream.256
----------------------------------------------
Double precision appears to have 16 digits of accuracy
Assuming 8 bytes per DOUBLE PRECISION word
----------------------------------------------
Array size = 1100791808
Offset = 255
The total memory requirement is **** MB
You are running each test 50 times
--
The *best* time for each test is used
*EXCLUDING* the first and last iterations
----------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds
----------------------------------------------------
Function Rate (MB/s) Avg time Min time Max time
Copy: 414572.5375 0.0472 0.0425 0.2525
Scale: 412108.3129 0.0436 0.0427 0.0571
Add: 485322.5573 0.0574 0.0544 0.1818
Triad: 488273.7434 0.0577 0.0541 0.2129
----------------------------------------------------
Solution Validates!
----------------------------------------------------
Mark Fahey
Center for Computational Sciences
Oak Ridge National Laboratory
Received on Mon Sep 29 08:03:15 2003
This archive was generated by hypermail 2.1.8 : Wed Oct 01 2003 - 07:21:30 CDT