A Parallel Numerical Solver Using Hierarchically Tiled Arrays

Download

index.pdf

Date

2011-01-01

Author

Brodman, James C.
Evans, G. Carl
Manguoğlu, Murat
Sameh, Ahmed
Garzaran, Maria J.
Padua, David

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

92
views

37
downloads

Solving linear systems is an important problem for scientific computing. Exploiting parallelism is essential for solving complex systems. and this traditionally involves writing parallel algorithms on top of a library such as MPI. The SPIKE family of algorithms is one well-known example of a parallel solver for linear systems. The Hierarchically Tiled Array data type extends traditional data-parallel array operations with explicit tiling and allows programmers to directly manipulate tiles. The tiles of the HTA data type map naturally to the block nature of many numeric computations, including the SPIKE family of algorithms. The higher level of abstraction of the HTA enables the same program to be portable across different platforms. Current implementations target both shared-memory and distributed-memory models. In this paper we present a proof-of-concept for portable linear solvers. We implement two algorithms from the SPIKE family using the HTA library. We show that our implementations of SPIKE exploit the abstractions provided by the HTA to produce a compact, clean code that can run on both shared-memory and distributed-memory models without modification. We discuss how we map the algorithms to HTA programs as well as examine their performance. We compare the performance of our HTA codes to comparable codes written in MPI as well as current state-of-the-art linear algebra routines.

Subject Keywords

Parallel solver, Tiling structure, Array operation, Clean code, Portable code

URI

https://hdl.handle.net/11511/46265

DOI

https://doi.org/10.1007/978-3-642-19595-2_4

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

A parallel sparse algorithm targeting arterial fluid mechanics computations Manguoğlu, Murat; Sameh, Ahmed H.; Tezduyar, Tayfun E. (2011-09-01) Iterative solution of large sparse nonsymmetric linear equation systems is one of the numerical challenges in arterial fluid-structure interaction computations. This is because the fluid mechanics parts of the fluid + structure block of the equation system that needs to be solved at every nonlinear iteration of each time step corresponds to incompressible flow, the computational domains include slender parts, and accurate wall shear stress calculations require boundary layer mesh refinement near the arteria...
A dual pair of optimization-based formulations for estimation and control Tuna, Sezai Emre (2015-01-01) A finite-horizon optimal estimation problem for discrete-time linear systems is formulated and solved. The formulation is a natural extension of that which yields a deadbeat observer. The resultant observer is the dual of the controller produced by the finite-horizon minimum energy control problem with terminal equality constraint. Nonlinear extensions of this dual pair are also considered and sufficient conditions are provided for stability and convergence.
Reevaluating Spectral Partitioning for Unsymmetric Matrices Oktay, Eda; Manguoğlu, Murat; Yücel, Hamdullah; Department of Scientific Computing (2020-9) Parallel solutions to scientific problems having graph representation require efficienttasks and partitioning data. In this thesis, various parallel graph partitioning algorithms are studied. While these algorithms are applicable to both directed and undirected graphs, we focus on the directed case whose matrix representations are sparse and unsymmetric arising in linear system of equations representing various application domains such as computational fluid dynamics and thermal problems. Strategies inspec...
A two dimensional euler flow solver on adaptive cartesian grids Siyahhan, Bercan; Aksel, Mehmet Haluk; Department of Mechanical Engineering (2008) In the thesis work, a code to solve the two dimensional compressible Euler equations for external flows around arbitrary geometries have been developed. A Cartesianmesh generator is incorporated to the solver. Hence the pre-processing can be performed together with the solution within a single code. The code is written in the C++ programming language and its object oriented capabilities have been exploited to save memory in the data structure developed. The Cartesian mesh is formed by dividing squares succe...
A New multi-threaded and recursive direct algorithm for parallel solution of sparse linear systems Bölükbaşı, Ercan Selçuk; Manguoğlu, Murat; Department of Computer Engineering (2013) Many of the science and engineering applications need to solve linear systems to model a real problem. Usually these linear systems have sparse coefficient matrices and thus require an effective solution of sparse linear systems which is usually the most time consuming operation. Circuit simulation, material science, power network analysis and computational fluid dynamics can be given as examples of these problems. With the introduction of multi-core processors, it became more important to solve sparse line...

Citation Formats

J. C. Brodman, G. C. Evans, M. Manguoğlu, A. Sameh, M. J. Garzaran, and D. Padua, “A Parallel Numerical Solver Using Hierarchically Tiled Arrays,” 2011, vol. 6548, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/46265.