A Class of Communication-avoiding Algorithms for Solving General Dense Linear Systems on CPU/GPU Parallel Machines
نویسندگان
چکیده
We study several solvers for the solution of general linear systems where the main objective is to reduce the communication overhead due to pivoting. We first describe two existing algorithms for the LU factorization on hybrid CPU/GPU architectures. The first one is based on partial pivoting and the second uses a random preconditioning of the original matrix to avoid pivoting. Then we introduce a solver where the panel factorization is performed using a communication-avoiding pivoting heuristic while the update of the trailing submatrix is performed by the GPU. We provide performance comparisons and tests on accuracy for these solvers on current hybrid multicore-GPU parallel machines.
منابع مشابه
Fast Cellular Automata Implementation on Graphic Processor Unit (GPU) for Salt and Pepper Noise Removal
Noise removal operation is commonly applied as pre-processing step before subsequent image processing tasks due to the occurrence of noise during acquisition or transmission process. A common problem in imaging systems by using CMOS or CCD sensors is appearance of the salt and pepper noise. This paper presents Cellular Automata (CA) framework for noise removal of distorted image by the salt an...
متن کاملA CPU-GPU hybrid approach for the unsymmetric multifrontal method
Multifrontal is an efficient direct method for solving large-scale sparse and unsymmetric linear systems. The method transforms a large sparse matrix factorization process into a sequence of factorizations involving smaller dense frontal matrices. Some of these dense operations can be accelerated by using a graphic processing unit (GPU). We analyze the unsymmetricmultifrontalmethod fromboth an ...
متن کاملGPU computing with Kaczmarz's and other iterative algorithms for linear systems
The graphics processing unit (GPU) is used to solve large linear systems derived from partial differential equations. The differential equations studied are strongly convection-dominated, of various sizes, and common to many fields, including computational fluid dynamics, heat transfer, and structural mechanics. The paper presents comparisons between GPU and CPU implementations of several well-...
متن کاملImplementation of the direction of arrival estimation algorithms by means of GPU-parallel processing in the Kuda environment (Research Article)
Direction-of-arrival (DOA) estimation of audio signals is critical in different areas, including electronic war, sonar, etc. The beamforming methods like Minimum Variance Distortionless Response (MVDR), Delay-and-Sum (DAS), and subspace-based Multiple Signal Classification (MUSIC) are the most known DOA estimation techniques. The mentioned methods have high computational complexity. Hence using...
متن کاملA novel cooperative accelerated parallel two-list algorithm for solving the subset-sum problem on a hybrid CPU-GPU cluster
Many parallel algorithms have recently been developed to accelerate solving the subset-sum problem on a heterogeneous CPU–GPU system. However, within each compute node, only one CPU core is used to control one GPU and all the remaining CPU cores are in idle state, which leads to a large number of CPU cores being wasted. In this paper, based on a cost-optimal parallel two-list algorithm, we prop...
متن کامل