MA Applied Mathematics with Graduate Certification in Scientific Computing, University of Pittsburgh, 1990. Numerical Analysis and Fluid Mechanics. Thesis: Optimization of Finite Element Codes. Application: Design of Reactor Core Cooling Water Systems.

BA Mathematics and Computer Science, University of Colorado, 1986. Applied Mathematics and Computer Science. Project: Statistical Evaluation of Instructor Effectiveness Using Optical Scanner Technology.

Senior Scientific Specialist, 2003-present
Pittsburgh Supercomputing Center, Carnegie Mellon University, Pittsburgh, PA

Terascale Computing Systems Liaison, 2001-2003
National Center for Supercomputing Applications, University of Illinois, Champaign, IL
Pittsburgh Supercomputing Center, Carnegie Mellon University, Pittsburgh, PA

Site Lead for Computational Structural Mechanics, 1997-2001
National Center for Supercomputing Applications, University of Illinois, Champaign, IL
Aeronautical Systems Center, Wright-Patterson Air Force Base, OH

Senior Scientific Software Developer, 1996-1997, Research Systems Programmer, 1994-1996, Senior Consultant, 1990-1994
Pittsburgh Supercomputing Center, Carnegie Mellon University, Pittsburgh, PA

Applications Engineer, 1987-1990
The Abelian Group, Pittsburgh, PA

Programmer/Analyst, 1986-1987, User Consultant and Recitation Leader, 1985-1986
University of Colorado, Department of Mathematics and Statistics, Denver, CO

Collaboration with MEDRAD, Inc. funded by the Pennsylvania Infrastructure Technology Alliance. Design optimization of a novel thromectomy device for treatment of blockages in arteriovenous grafts utilized in patients undergoing hemodialysis. Destruction of clots via mechanical means. 2007-present

Collaboration with the NSF Teragrid Project funded by the National Science Foundation. Serve as Grid Infrastructure Group Coordinator for the User Services Working Group. Organize formation of project teams, supervise support efforts, and collaborate as an embedded member of various research teams. 2005-present

Collaboration with Pittsburgh Supercomputing Center funded by the National Science Foundation's Partners in Academic Computing Initiative. Provision of consulting services for nationally recognized research groups including the Center for Analysis and Prediction of Storms (University of Oklahoma), the Theoretical Biophysics Group (University of Illinois), and the Applied Mathematics Group (Brown University). Specialist in porting issues and performance optimizations for various architectures such as the Cray XT3 and Compaq AlphaServer SC systems at PSC, and the IBM clusters at NCSA. 2001-2003

Collaboration with ANSYS, Inc. on a detailed evaluation of their Parallel Performance for ANSYS product. Focus on utilization and scalability of the Distributed Domain Solver releases 5.6 and 5.7 for Microsoft Windows/NT and Silicon Graphics IRIX clusters. Follow-on study of the 6.0 and 6.1 releases for Compaq Tru64 SC computing systems. 2000-2003

Computational Structural Mechanics supporting the DoD High Performance Computing Modernization Program. Planning, publication, educational outreach, student mentoring, software engineering (algorithm development, validation, performance testing), coupled physics problems and multidisciplinary optimization methods. Various collaborations with the Air Force Research Laboratory Materials and Air Vehicles Directorates, the Air Force Institute of Technology, the Army Research Laboratory, the University of Greenwich, Ohio State University, Rice University, Clark Atlanta University, and the University of Pittsburgh. 1997-2001

International Collaboration with the Universidad de Oriente, Venezuela. Parallelization of a multidisciplinary code used to model electrothermal processes associated with aluminum reduction cells (ARCs). An ARC is a type of industrial furnace used to separate aluminum from bauxite ore. The combination of extremes in temperatures and electrical currents coupled with a highly corrosive melt imply that ARCs are generally not well instrumented. Reduced scale models are used to estimate the thermal and electrical behaviors of production equipment. Because of the amount of damage caused when the content of an ARC device penetrates its refractory lining, accurate simulation has a well-defined value. Development of a three-dimensional transient model was completed. Work on OpenMP and explicit message passing versions of the new code initiated. Direction and assistance also provided to graduate students. 1998-2001

Collaboration with the Air Force Research Laboratory Materials Directorate. Analysis of solution algorithms employed by the Axisymmetric Damage Model (ADM) code. The ADM model and its derivatives serve to establish rigorous field theories for computing accurate stresses and energy release rates for fracture mechanics problems in composite bodies. A parallel version of ADM based on handwritten compiler directives was completed in June, 1998. Later work focused on the implementation of an alternative solution method capable of reducing the complexity of the kernel by an order of magnitude, while simultaneously eliminating the requirement for 128-bit floating point accuracy and range. 1997-2001

Collaboration with CEWES and NAVO Major Shared Resource Centers. Systems support and development of acceptance tests and benchmarking codes targeting miscellaneous high performance computers, networks, disk and tape drives. Constructed first 64-bit versions of PBLAS, BLACS, and ScaLAPACK for the Cray T3E. Patched data distribution and solver routines into a high-profile CFD code yielding a five-fold increase in throughput. 1997

Collaboration with Centric Engineering Systems. Ported parallel Spectrum application to the Cray T3E. Optimized single processor variants were ported to the J90 and T3E. Validation on both platforms was completed in November, 1996. A multiprocessor version for the J90 was subsequently completed and validated. Tests of the T3E parallel code revealed a problem with the PVM library component of Cray's Message Passing Toolkit. Nevertheless, the PVM port was capable of resolving about half of the vendor-supplied examples. Shortly after Centric's reorganization, the project was shelved. 1996-1997

Collaboration with the Applied Mathematics Group at the University of Pittsburgh. Implemented the Parallel Finite Element Method (PFEM) application, a research-level finite element model suitable for solving Dirichlet and Neumann boundary value problems on arbitrary domains in two dimensions. Optimized versions for the C90 and T3D were constructed in order to study the potential for integrating the algorithm into a standard finite element package on a distributed memory platform. The original source was also used to motivate the Summer Institute program at the Pittsburgh Supercomputing Center. 1995-1997

Design and Development of the Parallel Jacobi Eigensolver Library. The PJAC library was designed to provide a set of scalable eigensolvers for real/complex, symmetric/hermitian matrices. Problem sizes ranging from N=2 to 16384 were resolved on Cray MPP machines (T3D/T3E, 512 PEs). The algorithm employed is essentially a hybrid of Jacobi methods. As the processor count ranges from 1 to N/2, the macro-element Jacobi method transitions from a serial Jacobi method to the fine-grained Brent-Luk parallel model. Good performance and scaling were observed for a wide range of macro-element dimensions. 1995-1997

Design and Development of the Parallel File Input/Output Library. The PFIO library was designed to provide support for codes requiring optimal I/O performance. A pair of highly optimized I/O routines for moving data between Cray PVP and MPP memory and disk was developed. The PFIO and PFIO_IEEE libraries resulted directly from work on a common interface for our MPP eigensolvers. We were interested in feeding larger test problems into PJAC, and so a small amount of time was given to development of routines for performing asynchronous reads and writes of shared files at maximum bandwidth. The final T3E version incorporated user-level striping. 1995-1997

Engineering Software Maintenance. Installed, maintained, and provided user support for all fluid and solid mechanics software packages made available to users of the Pittsburgh Supercomputing Center, including ABAQUS, ANSYS, FIDAP, Fluent, LS-DYNA, NASTRAN, NEKTON, and PATRAN as well as a host of related pre- and post-processors. December 1995 to October 1997. Provided second-level support for a similar suite of applications to users of the Aeronautical Systems Center MSRC at Wright-Patterson Air Force Base, OH 1997-1999

Senior Consultant. Member of user services staff during the initial integration of massively parallel computers into the environment at Pittsburgh Supercomputing Center. Work centered on code optimization and technology transfer associated with Thinking Machines' CM-2 and CM-5 platforms and Cray's T3D and T3E systems. Pittsburgh Supercomputing Center, Pittsburgh, PA 1990-1994

Applications Engineer. Provision of technical services ranging from fluid systems design and production of engineering analysis tools to development and marketing of business-related database products. Also supported local area networks. The Abelian Group, Pittsburgh, PA 1987-1990

Programmer/Analyst. Conceived of and implemented an optical scanner interface program and complementary statistical analysis package used to conduct campus-wide faculty evaluations. Also served as computer laboratory consultant and recitation leader for the Mathematics and Computer Science departments. The University of Colorado, Denver, CO 1985-1987

R. Reddy and D. O'Neal, Adaptive MPI on the Teragrid, NSF Teragrid 2006 Conference, Indianapolis, IN. 2006

D. Weisser, N. Nystrom, S. Brown, J. Gardner, D. O'Neal, J. Urbanic, J. Lim, R. Reddy, R. Raymond, Y. Wang, and J. Welling, Performance of Applications on the Cray XT3, Cray User Group Conference, Lugano, Switzerland. 2006

D. O'Neal and S. Murgie, Evaluation of the Distributed Domain Solver, ANSYS Solutions, volume 4, number 3, Pittsburgh, PA. 2002

D. O'Neal and S. Murgie, ANSYS Benchmarking Project: Evaluation of the Domain Decomposition Solver, awarded Best of Session Technical Paper, Proceedings of the ANSYS 2002 Conference, Pittsburgh, PA. 2002

D. O'Neal, R. Luczak, N. Pagano, G. Schoeppner, H. Brown III, and G. Tandon, MDM Project: Analysis of the Micromechanics Damage Model Suite, Proceedings of MAPINT 2000, 4th Symposium on Multidisciplinary Applications and Interoperable Computing, Dayton, OH. 2000

D. O'Neal, R. Luczak, and M. White, CAPTools Project Final Report: Evaluation and Application of the Computer Aided Parallelisation Tools, Proceedings of the DoD High Performance Computing Modernization Program Users Group Conference, Albuquerque, NM. 2000

D. O'Neal and R. Reddy, Solving Symmetric Eigenvalue Problems, Industrial Strength Parallel Computing, Chapter 17, Morgan Kauffman Publishers Inc, San Francisco, CA. 1999

D. O'Neal and J. Urbanic, On Microprocessors, Memory Hierarchies, and Amdahl's Law, DoD HPC Users Group Conference, Monterey, CA. 1999

D. O'Neal and R. Luczak, CAPTools Project: Evaluation and Application of the Computer Aided Parallelisation Tools, DoD HPC Users Group Conference, Monterey, CA. 1999

D. O'Neal, W. Layton, and N. Troyani, Aluminum Reduction Cells and Electrothermal Models, Project abstract and proposal for supercomputing resources, Pittsburgh Supercomputing Center, Pittsburgh, PA. 1998

D. O'Neal. N. Pagano, G. Shoeppner, and G. Tandon, A Case Study of Multithread Porting Issues: The Axisymmetric Damage Model, DoD HPC Users Group Conference, Rice University, Houston, TX. 1998

N. Nystrom and D. O'Neal, Applications Performance on the Cray T3E and T3E-900, Cray User Group Proceedings, Spring Conference, San Jose, CA. 1997

D. O'Neal and J. Urbanic, On Performance and Efficiency: Cray Architectures, Conference Proceedings, Parallel Computing 1997, Ohio Supercomputer Center, Columbus, OH. 1997

D. O'Neal, High Performance Engineering, Westinghouse Corporate Symposium, Westinghouse Energy Center, Pittsburgh, PA. 1996

D. O'Neal and R. Reddy, The Parallel Finite Element Method, Cray User Group Proceedings, Fall Conference, Denver, CO. 1995

D. O'Neal and R. Reddy, Solving Symmetric Eigenvalue Problems on Distributed Memory Machines, Cray User Group Proceedings, Fall Conference, Tours, France. 1994

D. O'Neal and R. Reddy, Constructing and Managing a Group Programming Environment, PSC News Update #2, v. 6, Pittsburgh Supercomputing Center, Pittsburgh, PA. 1993

D. O'Neal, Reactor Core Cooling Water Systems, MA thesis project, University of Pittsburgh, Pittsburgh, PA. 1990

D. O'Neal, Optimization of Finite Element Codes, J. of Sci. Comp. 10, p. 36-52, Plenum Press, New York, NY. 1989

D. O'Neal and A. Sunmonu, On Finite Element Codes, University of Pittsburgh, Pittsburgh, PA. 1988

On Grid Computing: The Teragrid Production Facility, Super Computing Science Consortium (SC)², Waynesburg College, Waynesburg, PA. 2006

On Performance and Efficiency: Cray Architectures, Pittsburgh Supercomputing Center, MPP Workshop Series. 1997

Scientific Applications: The Parallel Finite Element Method, Pittsburgh Supercomputing Center, MPP Workshop Series. 1996-1997

Scientific Applications: Parallel Jacobi Methods, Pittsburgh Supercomputing Center, MPP Workshop Series. 1995-1996

Introduction to Parallel Computing, Pittsburgh Supercomputing Center, C90/J90 Workshop Series. 1993-1997

Code Analysis, Pittsburgh Supercomputing Center, C90/J90 Workshop Series. 1993-1997