Domain-specific virtual processors as a portable programming and execution model for parallel computational workloads on modern heterogeneous high-performance computing architectures
Abstract
We advocate domain-specific virtual processors (DSVP) as a portability layer for expressing and executing domain-specific computational workloads on modern heterogeneous HPC architectures, with applications in quantum chemistry. Specifically, in this article we extend, generalize and better formalize the concept of a domain-specific virtual processor as applied to scientific high-performance computing. In particular, we introduce a system-wide recursive (hierarchical) hardware encapsulation mechanism into the DSVP architecture and specify a concrete microarchitectural design of an abstract DSVP from which specialized DSVP implementations can be derived for specific scientific domains. Furthermore, we demonstrate, an example of a domain-specific virtual processor specialized to numerical tensor algebra workloads, which is implemented in the ExaTENSOR library developed by the author with a primary focus on the quantum many-body computational workloads on large-scale GPU-accelerated HPC platforms.
- Authors:
-
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Publication Date:
- Research Org.:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
- Sponsoring Org.:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- OSTI Identifier:
- 1509576
- Grant/Contract Number:
- AC05-00OR22725
- Resource Type:
- Accepted Manuscript
- Journal Name:
- International Journal of Quantum Chemistry
- Additional Journal Information:
- Journal Volume: 119; Journal Issue: 12; Journal ID: ISSN 0020-7608
- Publisher:
- Wiley
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; heterogeneous node architecture; high‐performance computing; numerical tensor algebra; scientific computing; virtual machine
Citation Formats
Lyakh, Dmitry I. Domain-specific virtual processors as a portable programming and execution model for parallel computational workloads on modern heterogeneous high-performance computing architectures. United States: N. p., 2019.
Web. doi:10.1002/qua.25926.
Lyakh, Dmitry I. Domain-specific virtual processors as a portable programming and execution model for parallel computational workloads on modern heterogeneous high-performance computing architectures. United States. https://doi.org/10.1002/qua.25926
Lyakh, Dmitry I. Tue .
"Domain-specific virtual processors as a portable programming and execution model for parallel computational workloads on modern heterogeneous high-performance computing architectures". United States. https://doi.org/10.1002/qua.25926. https://www.osti.gov/servlets/purl/1509576.
@article{osti_1509576,
title = {Domain-specific virtual processors as a portable programming and execution model for parallel computational workloads on modern heterogeneous high-performance computing architectures},
author = {Lyakh, Dmitry I.},
abstractNote = {We advocate domain-specific virtual processors (DSVP) as a portability layer for expressing and executing domain-specific computational workloads on modern heterogeneous HPC architectures, with applications in quantum chemistry. Specifically, in this article we extend, generalize and better formalize the concept of a domain-specific virtual processor as applied to scientific high-performance computing. In particular, we introduce a system-wide recursive (hierarchical) hardware encapsulation mechanism into the DSVP architecture and specify a concrete microarchitectural design of an abstract DSVP from which specialized DSVP implementations can be derived for specific scientific domains. Furthermore, we demonstrate, an example of a domain-specific virtual processor specialized to numerical tensor algebra workloads, which is implemented in the ExaTENSOR library developed by the author with a primary focus on the quantum many-body computational workloads on large-scale GPU-accelerated HPC platforms.},
doi = {10.1002/qua.25926},
journal = {International Journal of Quantum Chemistry},
number = 12,
volume = 119,
place = {United States},
year = {Tue Mar 12 00:00:00 EDT 2019},
month = {Tue Mar 12 00:00:00 EDT 2019}
}
Web of Science
Works referenced in this record:
Numerical linear algebra on emerging architectures: The PLASMA and MAGMA projects
journal, July 2009
- Agullo, Emmanuel; Demmel, Jim; Dongarra, Jack
- Journal of Physics: Conference Series, Vol. 180
Automatic Task-Based Code Generation for High Performance Domain Specific Embedded Language
journal, March 2015
- Tran Tan, Antoine; Falcou, Joel; Etiemble, Daniel
- International Journal of Parallel Programming, Vol. 44, Issue 3
Exploiting GPUs with the Super Instruction Architecture
journal, August 2014
- Jindal, Nakul; Lotrich, Victor; Deumens, Erik
- International Journal of Parallel Programming, Vol. 44, Issue 2
Parallel implementation of electronic structure energy, gradient, and Hessian calculations
journal, May 2008
- Lotrich, V.; Flocke, N.; Ponton, M.
- The Journal of Chemical Physics, Vol. 128, Issue 19
Heterogeneous parallel and distributed computing
journal, December 1999
- Sunderam, V. S.; Geist, G. A.
- Parallel Computing, Vol. 25, Issue 13-14
The Open Community Runtime: A runtime system for extreme scale computing
conference, September 2016
- Mattson, Timothy G.; Cledat, Romain; Cave, Vincent
- 2016 IEEE High Performance Extreme Computing Conference (HPEC)
General atomic and molecular electronic structure system
journal, November 1993
- Schmidt, Michael W.; Baldridge, Kim K.; Boatz, Jerry A.
- Journal of Computational Chemistry, Vol. 14, Issue 11, p. 1347-1363
Replicated Computational Results Certified BLIS: A Framework for Rapidly Instantiating BLAS Functionality
journal, June 2015
- Zee, Field G. VAN; Geijn, Robert A. VAN DE
- ACM Transactions on Mathematical Software, Vol. 41, Issue 3
NWChem: A comprehensive and scalable open-source solution for large scale molecular simulations
journal, September 2010
- Valiev, M.; Bylaska, E. J.; Govind, N.
- Computer Physics Communications, Vol. 181, Issue 9, p. 1477-1489
Many – Body Methods in Chemistry and Physics
book, January 2009
- Shavitt, Isaiah; Bartlett, Rodney J.
The ELPA library: scalable parallel eigenvalue solutions for electronic structure theory and computational science
journal, May 2014
- Marek, A.; Blum, V.; Johanni, R.
- Journal of Physics: Condensed Matter, Vol. 26, Issue 21
A massively parallel tensor contraction framework for coupled-cluster computations
journal, December 2014
- Solomonik, Edgar; Matthews, Devin; Hammond, Jeff R.
- Journal of Parallel and Distributed Computing, Vol. 74, Issue 12
Super instruction architecture of petascale electronic structure software: the story
journal, November 2010
- Lotrich, V. F.; Ponton, J. M.; Perera, A. S.
- Molecular Physics, Vol. 108, Issue 21-23
Works referencing / citing this record:
Establishing the Quantum Supremacy Frontier with a 281 Pflop/s Simulation
text, January 2019
- Villalonga, Benjamin; Lyakh, Dmitry; Boixo, Sergio
- arXiv