36. NUMERICAL SOFTWARE MAINTENANCE  

The Office of Advanced Scientific Computing Research has been fully or partially responsible for funding the research and development (R&D) of a wide range of robust, high-quality numerical algorithms for scientific computation.  These include the development of libraries such as EISPACK, LINPACK, LAPACK, ScaLAPACK, ARPACK, CLAWPACK, PETSc, TAO, CHOMBO, ebCHOMBO, SALSA, MPSALSA, LOCA, HYPRE, SuperLU, FronTier, and many others.  However, a number of critical issues must still be resolved in order to ensure that the value of the software is maintained and that the large R&D investment is maximized. 

a. Numerical Software Maintenance, Versioning, and Distribution—Grant applications are sought for the development of technologies to provide:  enhanced user interfaces; distribution support; maintenance activities such as collecting and tracking bug reports, and fixing bugs; and portability across platforms (including porting to new computational architectures).  Grant applications are also sought to:  (1) develop new maintenance and distribution mechanisms to ensure that updated scientific libraries are subjected to validation and verification testing; (2) implement formal tracking mechanisms for bug reports, bug fixes, and update notification for a wide range of scientific algorithm libraries; (3) develop and maintain mechanisms for providing cost effective portability of scientific libraries across a wide range of computer architectures, from desktop systems to massively parallel leadership-class supercomputers; (4) develop and maintain high-quality user documentation for each component of scientific software, including advice on domains of applicability for each module; and (5) develop comprehensive email- or Web-based user support services for scientific libraries. 

Questions – contact Thomas Ndousse-Fetter (tndousse@science.doe.gov

b.  Scaling Mathematical Tools and Libraries to PetascaleThe DOE Office of Science has entered into the era of petascale computer science – marked by computers that operate a 1000 times faster than today’s teraflop computers.  Petascale computing will enable the production of scientific simulation data about complex natural phenomena, on a scale not possible just a few years ago.  Critical issues that must be resolved to enable science at the petascale computing is scaling existing mathematical libraries and tools to take full advantage of the petascale computing.  Grant applications are sought from investigators that will collaborate with domain scientists to scale existing mathematical libraries, solvers, and tools to work efficiently in petascale computers at the National Leadership Facilities at Oak Ridge National Laboratory, Argonne National Laboratory, and National Energy Research Scientific Computing Center.

Questions – contact Thomas Ndousse-Fetter (tndousse@science.doe.gov

References:

1.      Anderson, E., et al., “LAPACK Users' Guide,” 2nd ed., Philadelphia, PA:  Society for Industrial and Applied Mathematics (SIAM), 1995.  (ISBN:  0-89871-345-5)

2.      Dongarra, J. and Walker, D., “Software Libraries for Linear Algebra Computations on High Performance Computers,” SIAM Review, 37: 151-180, 1995.  (ISSN:  0036-1445)

3.      Dongarra, J. J., et al., “Algorithm 679:  A Set of Level 3 Basic Linear Algebra Subprograms,” ACM [Association for Computing] Transactions on Mathematical Software, 16(1): 8-28, March 1990.  (ISSN:  0098-3500)

4.      Dongarra, J. J., et al., Algorithm 656:  An Extended Set of FORTRAN Basic Linear Algebra Subroutines,” ACM Transactions on Mathematical Software, 14(1): 18-32, March 1988.  (ISSN:  0098-3500)

5.      Geist, A., et al., eds., “PVM:  Parallel Virtual Machine. A Users' Guide and Tutorial for Networked Parallel Computing,” Cambridge, MA:  MIT Press, 1994.  (ISBN:  0262571080)

6.      Pollicini, A. A., “Using Toolpack Software Tools,” Kluwer Academic Publishers, 1989.  (ISBN:  0-7923-0033-5)

7.      Blackford, L. S., et al., “The ScaLAPACK Users Guide,” Philadelphia, PA:  SIAM, 1997.  (ISBN:  0-89871-397-8)

8.      Smith, B. T., et al., “Matrix Eigensystem Routines,” EISPACK Guide Lecture Notes in Computer Science, 2nd ed., Vol. 6, Springer-Verlag, 1976.  (ISBN:  0-38707-546-1)

9.      Lehoucq, R. B., et al., “ARPACK Users Guide:  Solution of Large-Scale Eigenvalue Problems with Implicitly Restarted Arnoldi Methods,” Philadelphia, PA:  SIAM, 1998.  (ISBN:  0-89871-407-9)

10.  Balay, S., et al., “Efficient Management of Parallelism in Object Oriented Numerical Software Libraries,” in Modern Software Tools in Scientific Computing, pp. 163-202, Birkhauser Press, 1997.  (ISBN:  0-8176-3974-8)

11.  Balay, S., et al., “PETSc Users Manual,” Argonne National Laboratory, 2002.  (Report No. ANL-95/11 - Rev. 2.1.6)(Full text available at:  http://www-unix.mcs.anl.gov/petsc/petsc-2/snapshots/petsc-current/docs/manual.pdf)

12.  Benson, S., et al., “TAO Users Manual,” Technical Report, Argonne National Laboratory, August 2004.  (Report No. ANL/MCS-TM-242-Revision 1.7)(Full text available at:  http://www-unix.mcs.anl.gov/tao/docs/manual/manual.html)

13.  Shadid, J., et al., “MPSalsa Version 1.5:  A Finite Element Computer Program for Reacting Flow Problems:  Part 1 – Theoretical Development,” Technical Report, Sandia National Laboratories, 1998.  (Report No. SAND98-2864) (Full text available at:  http://www.osti.gov/bridge/servlets/purl/2641-t7isU8/webviewable/2641.PDF)

 

Return to the Complete List of Topics

 

Program Information, Instructions and Requirements  |  Technical Topic Descriptions  |  Download Program Information  | Download Technical Topics |