Generalized asynchronous iterations
Proc. of the conference on algorithms and hardware for parallel processing on CONPAR 86
Journal of Computational and Applied Mathematics - Special issue on numerical analysis 2000 Vol. III: linear algebra
A Parallel Krylov-Type Method for Nonsymmetric Linear Systems
HiPC '01 Proceedings of the 8th International Conference on High Performance Computing
Iterative Methods for Sparse Linear Systems
Iterative Methods for Sparse Linear Systems
Parallel Scientific Computing in C++ and MPI
Parallel Scientific Computing in C++ and MPI
Architectural core salvaging in a multi-core processor for hard-error tolerance
Proceedings of the 36th annual international symposium on Computer architecture
International Journal of High Performance Computing Applications
The International Exascale Software Project roadmap
International Journal of High Performance Computing Applications
Cooperative Application/OS DRAM fault recovery
Euro-Par'11 Proceedings of the 2011 international conference on Parallel Processing - Volume 2
A Block-Asynchronous Relaxation Method for Graphics Processing Units
IPDPSW '12 Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum
Self-stabilizing iterative solvers
ScalA '13 Proceedings of the Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems
Hi-index | 0.00 |
In this paper, we analyze the potential of asynchronous relaxation methods on Graphics Processing Units (GPUs). We develop asynchronous iteration algorithms in CUDA and compare them with parallel implementations of synchronous relaxation methods on CPU- or GPU-based systems. For a set of test matrices from UFMC we investigate convergence behavior, performance and tolerance to hardware failure. We observe that even for our most basic asynchronous relaxation scheme, the method can efficiently leverage the GPUs computing power and is, despite its lower convergence rate compared to the Gauss-Seidel relaxation, still able to provide solution approximations of certain accuracy in considerably shorter time than Gauss-Seidel running on CPUs- or GPU-based Jacobi. Hence, it overcompensates for the slower convergence by exploiting the scalability and the good fit of the asynchronous schemes for the highly parallel GPU architectures. Further, enhancing the most basic asynchronous approach with hybrid schemes-using multiple iterations within the ''subdomain'' handled by a GPU thread block-we manage to not only recover the loss of global convergence but often accelerate convergence of up to two times, while keeping the execution time of a global iteration practically the same. The combination with the advantageous properties of asynchronous iteration methods with respect to hardware failure identifies the high potential of the asynchronous methods for Exascale computing.