CAPO - Computer-Aided Parallelizer and Optimizer

CAPO (Computer-Aided Parallelizer and Optimizer)

Overview

CAPO (Computer-Aided Parallelizer and Optimizer) automates the insertion of compiler directives to facilitate parallel processing on shared memory parallel (SMP) machines. While CAPO is currently integrated seamlessly into CAPTools (developed at the University of Greenwich, now marketed as ParaWise) CAPO is independently developed at NASA Ames Research Center as one of the components for the Legacy Code Modernization (LCM) project. The current version takes serial FORTRAN programs, performs interprocedual data dependence analysis, and generates OpenMP directives. Due to the widely support of the OpenMP standard, the generated OpenMP codes can potentially run on a wide range of SMP machines.

Approach

The success of CAPO relies on accurate interprocedual data dependence information which is currently provided by CAPTools. CAPO generates compiler directives in three stages:

identification of parallel loops in the outer-most level,
construction of parallel regions around parallel loops and optimization of parallel regions, and
insertion of directives with automatic identification of private, reduction, induction, and shared variables.

Attempts have also been made to identify potential pipeline parallelism (implemented with point-to-point synchronization). Although directives are generated automatically, user interaction with the tool is still important for producing good parallel codes. A comprehensive graphical user interface is included for user to interact with the parallelization process.

CAPO Graphical User Interface

Here is the main GUI (Directives Browser) for CAPO. [Directives Browser]

Result

Here is a test case of CAPO on parallelizing a large CFD code.

Application:: OVERFLOW (~100K lines, 870 subroutines)
Dependence analysis:: 3 days on an SGI R10K (175MHz) workstation
Post analysis and code testing:: another 3 days
Testing platform:: Origin2000, 195MHz
Testing cases:: single zone, multiple sizes
Conclusion:: CAPO is capable of generating parallel codes with directives quickly and good performance can be achieved with small amount of user interaction.

Some Documents and Publications

"The Code Parallelization with CAPO - A User Manual," Haoqiang Jin, Michael Frumkin and Jerry Yan, NAS Technical Report NAS-01-008, NASA Ames Research Center, August 2001. (try link 2 if the first one failed).
"Automatic Generation of OpenMP Directives and Its Application to Computational Fluid Dynamics Codes," H. Jin, M. Frumkin and J. Yan, presented in the Workshop on OpenMP Experiences and Implementations (WOMPEI), in Proceedings of the Third International Symposium on High Performance Computing (ISHPC), Tokyo, Japan, October 2000.
"The Semi-Automatic Parallelisation of Scientific Application Codes Using a Computer Aided Parallelisation Toolkit," C. Ierotheou, S. Johnson, P. Leggett, M. Cross, E. Evans (Univ. of Greenwich), H. Jin, M. Frumkin and J. Yan (NASA Ames), presented at the Workshop on OpenMP: Application and Tools (WOMPAT), San Diego, CA, 2000; in Special Issue of Scientific Programming Journal, Vol. 9, No.2+3, 2001, p163-173.
"Automatic Multilevel Parallelization Using OpenMP," H. Jin, G. Jost, J. Yan (NASA Ames), E. Ayguade, M. Gonzalez, and X. Martorell (CEPBA), presented at the European Workshop on OpenMP (EWOMP2001), Barcelona, Spain, September 2001; Scientific Programming, Vol. 11, No. 2 (2003) 177-190.
"Interfacing Computer Aided Parallelization and Performance Analysis" G. Jost, H. Jin (NASA Ames), J. Labarta, J. Gimenez (CEPBA), presented in the International Conference on Computational Science 2003 (ICCS2003), special section on Tools for Programming Development and Analysis in Computational Science Melbourne, Australia, June 2003.
"Experience on the Parallelization of a Cloud Modeling Code Using Computer-Aided Tools" Haoqiang Jin, Gabriele Jost (NASA Ames), Dan Johnson and Wei-Kuo Tao (NASA GSFC), NAS Technical Report NAS-03-006, NASA Ames Research Center, March 2003.
"Automatic Relative Debugging of OpenMP Programs" G. Matthews, R. Hood, H. Jin (NASA Ames), S. Johnson, and C. Ierotheou (Univ. of Greenwich), in the Proceedings of the European Workshop on OpenMP (EWOMP2003), Aachen, Germany, Sept. 22-26, 2003. (NAS-03-014)

xview-3.2.tar for Linux.
CAPO examples (version 1.1).

Point of Contact

For further information on CAPO, please contact:

hjin@nas.nasa.gov

capo@nas.nasa.gov

To obtain a beta-test copy of CAPO for a 6-month trial period, fill out the Non-Disclosure Agreement form (MS Word 23KB, PDF 4KB) and fax to the address indicated at the bottom of the form, in addition, cc a copy to 1-650-604-3957.

For US entities, the distribution of CAPO is available through the Software Usage Agreement with NASA.

Updated: Oct. 20, 2002
WebWork: Haoqiang Jin
NASA Responsible Official: Byran Biegel