Abstract
In this paper we evaluate the possibility of using mixed precision algorithms on different hardware platforms to obtain energy-efficient solvers for linear systems of equations. Our test-cases arise in the context of computational fluid dynamics.
Therefore, we analyze the energy efficiency of common cluster nodes and a hybrid, GPU-accelerated cluster node, when applying a linear solver, that can benefit from the use of different precision formats.
We show the high potential of hardware-aware computing in terms of performance and energy efficiency.
Similar content being viewed by others
References
Lampe F (2010) Mit neuen IT-Technologien Energieeffizienz erreichen, die Umwelt schonen und Kosten sparen, 1st edn. Vieweg+Teubner/GWV Fachverlage GmbH, Wiesbaden
ExaScale Computing Study (2008) Technology challenges in achieving ExaScale systems, DARPA/IPTO, 297 p
Anzt H, Rocker B, Heuveline V (2010) Mixed precision error correction methods for linear systems: convergence analysis based on Krylov subspace methods. In: Proceedings of PARA 2010. State of the art in scientific and parallel computing, 11 p
Göddeke D, Strzodka R, Turek S (2007) Performance and accuracy of hardware- oriented native-, emulated- and mixed-precision solvers in FEM simulations. Int J Parallel Emergent Distributed Syst 22:1744–1760. Special issue: applied parallel computing
Göddeke D, Strzodka R (2008) Performance and accuracy of hardware-oriented native-, emulated- and mixed-precision solvers in FEM simulations (Part 2: Double precision GPUs). Technical report, Fakultät für Mathematik TU Dortmund, 9 p
Buttari A, Dongarra J, Langou J, Langou J, Luszcek P, Kurzak J (2007) Mixed precision iterative refinement techniques for the solution of dense linear systems. Int J High Perform Comput Appl 21(4):457–466
Baboulin M, Buttari A, Dongarra J, Langou J, Langou J, Luszcek P, Kurzak J, Tomov S (2009) Accelerating scientific computations with mixed precision algorithms. Comput Phys Commun 180:2526–2533
Anzt H, Rocker B, Heuveline V (2010) An error correction solver for linear systems: evaluation of mixed precision implementations. In: Proceedings of VECPAR 2010. 9th international meeting on high performance computing for computational science, 8 p
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Anzt, H., Rocker, B. & Heuveline, V. Energy efficiency of mixed precision iterative refinement methods using hybrid hardware platforms. Comput Sci Res Dev 25, 141–148 (2010). https://doi.org/10.1007/s00450-010-0124-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00450-010-0124-2