A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Filters
CPU-GPU hybrid bidiagonal reduction with soft error resilience
2013
Proceedings of the Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - ScalA '13
Preserved Fulltext
In this paper, we present a design of a bidiagonal reduction algorithm that is resilient to soft errors, and we also describe its implementation on hybrid CPU-GPU architectures. ...
The tests were performed on a Sandy Bridge CPU coupled with an NVIDIA Kepler GPU. ...
Our fault tolerant bidiagonal reduction algorithm employs reverse computation and algorithm-based fault tolerance to detect, locate, and correct soft errors in the bidiagonal reduction on CPU-GPU hybrid ...
doi:10.1145/2530268.2530270
dblp:conf/sc/JiaLBD13
fatcat:fnzowqdodzhz5bl57mwkeujere
Algorithm-Based Fault Tolerance for Two-Sided Dense Matrix Factorizations
2015
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
We also designed algorithm-based fault tolerant algorithms for the CPU-GPU vii hybrid Hessenberg reduction algorithm and the CPU-GPU hybrid bidiagonal reduction algorithm. ...
Solving EVP and SVP numerically involves two-sided matrix factorizations: the Hessenberg reduction, the tridiagonal reduction, and the bidiagonal reduction. ...
This work also addresses soft errors in the Hessenberg reduction algorithm and the bidiagonal reduction algorithms on CPU-GPU hybrid algorithms. ...
fatcat:cs2f6ueb2bdkherfcy3udzn6xq
Welcome from the SHAXC-2 Organizing Committee Aims and Scope SHAXC-2 Organizing Committee
2014
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
The challenges of the exascale include: reducing communication, reducing synchronization, increasing concurrency to exploit specialized manycore and GPU processing elements, and incorporating resilience ...
into the algorithms, themselves, rather than leaving the entire burden of resilience to the hardware. ...
He is interested in optimizing dense linear algebra for hybrid-distributed memory systems equipped with GPUs. ...
fatcat:rdvlyeogwjbnzk6p2j6lzm2see
Scalability and fault tolerance of ABFT methods for dense matrix multiplication
2018
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
As an essential basis for the results DPLASMA, a highly optimized library for distributed hybrid systems was used. ...
Therefore, the need for exhaustive fault detection and error correction algorithms become increasingly important over the last few years. ...
The scope for singular values is defined by reducing to condensed form where it can be done either by reduction to band-bidiagonal or reduction to proper bidiagonal. ...
doi:10.25365/thesis.51789
fatcat:m7t2bxc3sfc25ckuuo5npbgr2q