Mike Heroux Sandia National Laboratories
Mike Heroux Michael A. Heroux
Distinguished Member of the Technical Staff
Scalable Algorithms Department, 1416
Computation, Computers, Information and Mathematics Center

Email: maherou at sandia.gov
Phone Number: 1-320-845-7695 (1-505-845-7695 when in ABQ)
Fax Number: 1-320-845-7846
Office: 18125 Kreigle Lake Road, Avon, MN 56310

++ Upcoming Conference and Workshop Information:
4th International Conference on Supercomputing June 1-4, 2010 Tsukuba, Japan
2010 International Workshop on OpenMP (IWOMP 2010) June 14-16, 2010, Tsukuba, Japan
SPEEDUP Workshop on High-Performance Computing. September 6-7, 2010, Zurich, Switzerland.
International Exascale Software Project Meeting, September 2010, Maui, Hawaii (tentative)
++ Recent or Frequently-requested Technical Reports and Papers:
On the Path to Exascale
Factors Impacting Performance of Multithreaded Sparse Triangular Solve
Sustainable CSE Software Engineering and Support: Lessons Learned from the Trilinos Project
Improving Performance via Mini-applications (Mantevo Overview)
A Light-weight API for Portable Multicore Programming (PDP 2010, accepted)
Software Challenges for Extreme Scale Computing (International Journal of High Performance Computing, to appear)
Implications of System Errors in the Context of Numerical Accuracy
Barely Sufficient Software Engineering: 10 Practices to Improve Your CSE Software (SECSE 2009, ICSE 2009)
Design issues for numerical libraries on scalable multicore architectures (SciDAC 2008 paper)
The Changing Scope of Trilinos
Trilinos Overview (ACM TOMS Paper)
Parallel segregated Schur complement methods for fluid density functional theories (SIAM SISC paper)
The Trilinos Software Lifecycle Model (Proceedings of ICSE 2007)
Improving the Development Process for CSE Software (Proceedings of PDP2007)
A Solver-Independent API for multi-DOF Applications using Trilinos (IJCSE paper)
PyTrilinos Overview (ACM TOMS Paper)
++ Recent or Frequently-requested Presentations:
Trilinos Tutorial (2009 Trilinos User Group Meeting). PDF Version here.
Scalable Computing Challenges: An Overview
Scalable Algorithms for 1M Cores: What Might and Might Not Work, and Why
Next-generation Scalable Applications: When MPI-only is not enough, Sandia CSRI Workshop, Santa Fe, NM, June 3-5, 2008
Been There, Done That: Lessons Learned from SMP Computing (Multicore/Manycore Workshop 2007 Talk)
Performance Issues for Sparse Iterative Solvers Talk
Presentation from GP2: Workshop on General Purpose Computing on Graphics Processors, Los Angeles, CA August 7-8, 2004
Trilinos Tutorial from Seventh ACTS Workshop, LBL, Berkeley, CA August 22-25, 2006
++ Software Downloads:
Trilinos Download Page
Mantevo Micro-Application Download Page