Filters








1,098 Hits in 4.1 sec

Algorithm-based fault tolerance for dense matrix factorizations

Peng Du, Aurelien Bouteiller, George Bosilca, Thomas Herault, Jack Dongarra
2012 SIGPLAN notices  
This paper proposes a new hybrid approach, based on Algorithm-Based Fault Tolerance (ABFT), to help matrix factorizations algorithms survive fail-stop failures.  ...  The fault-tolerant algorithms derived from this hybrid solution is applicable to a wide range of dense matrix factorizations, with minor modifications.  ...  Algorithm Based Fault Tolerance Background The most well-known fault-tolerance technique for parallel applications, checkpoint-restart (C/R), encompasses two categories, the system and application level  ... 
doi:10.1145/2370036.2145845 fatcat:aj5ivcrf25f4reulnpy6cpcdiy

Algorithm-based fault tolerance for dense matrix factorizations

Peng Du, Aurelien Bouteiller, George Bosilca, Thomas Herault, Jack Dongarra
2012 Proceedings of the 17th ACM SIGPLAN symposium on Principles and Practice of Parallel Programming - PPoPP '12  
This paper proposes a new hybrid approach, based on Algorithm-Based Fault Tolerance (ABFT), to help matrix factorizations algorithms survive fail-stop failures.  ...  The fault-tolerant algorithms derived from this hybrid solution is applicable to a wide range of dense matrix factorizations, with minor modifications.  ...  Algorithm Based Fault Tolerance Background The most well-known fault-tolerance technique for parallel applications, checkpoint-restart (C/R), encompasses two categories, the system and application level  ... 
doi:10.1145/2145816.2145845 dblp:conf/ppopp/DuBBHD12 fatcat:cyc73fwdtvhhve7gjzi6vyc7ne

Algorithm-Based Fault Tolerance for Dense Matrix Factorizations, Multiple Failures and Accuracy

Aurelien Bouteiller, Thomas Herault, George Bosilca, Peng Du, Jack Dongarra
2015 ACM Transactions on Parallel Computing  
This paper proposes a new hybrid approach, based on Algorithm-Based Fault Tolerance (ABFT), to help matrix factorizations algorithms survive fail-stop failures.  ...  The fault-tolerant algorithms derived from this hybrid solution is applicable to a wide range of dense matrix factorizations, with minor modifications.  ...  ALGORITHM BASED FAULT TOLERANCE BACKGROUND The most well-known fault-tolerance technique for parallel applications, checkpointrestart (C/R), encompasses two categories, the system and application level  ... 
doi:10.1145/2686892 fatcat:yu4orwb2uncgxbdzyrs6xw5z2e

Page 350 of IEEE Transactions on Computers Vol. 52, Issue 3 [page]

2003 IEEE Transactions on Computers  
Feng, “Algorithm Based Fault Tolerance for |FEE Computer-Society the ACM. sched Matrix Inversion with Maximum Pivoting,” J.  ...  Abraham, “Algorithm-Based Fault Toler- ance for Matrix Operations,” [EEE Trans. Computers, vol. 33, pp 518-528, 1984 [11] M. Joseph and P.  ... 

Scalable matrix inversion using MapReduce

Jingen Xiang, Huangdong Meng, Ashraf Aboulnaga
2014 Proceedings of the 23rd international symposium on High-performance parallel and distributed computing - HPDC '14  
In this paper, we present a scalable and efficient technique for matrix inversion in MapReduce.  ...  We present a technique for computing the LU decomposition and the matrix inverse using a pipeline of MapReduce jobs. We also present optimizations of this technique in the context of Hadoop.  ...  CONCLUSIONS AND FUTURE WORK We presented a scalable and fault tolerant algorithm for matrix inversion using MapReduce.  ... 
doi:10.1145/2600212.2600220 dblp:conf/hpdc/XiangMA14 fatcat:htqbccqjpnflxmfkjahaagtws4

Algorithm-Based Fault Tolerance for Matrix Operations

Kuang-Hua Huang, Abraham
1984 IEEE transactions on computers  
The computation tasks within an algorithm are appropriately distributed among multiple computation units for fault tolerance.  ...  Algorithm-based fault tolerance schemes are proposed to detect and correct errors when matrix operations such as addition, multiplication, scalar product, LU-decomposition, and transposition are performed  ...  ALGORITHM-BASED FAULT TOLERANCE This section describes the basic ideas behind the algorithm-based fault tolerance technique.  ... 
doi:10.1109/tc.1984.1676475 fatcat:esqcnwz4nff7xbbxbaisezj2jm

Spark-Based Large-Scale Matrix Inversion for Big Data Processing

Jun Liu, Yang Liang, Nirwan Ansari
2016 IEEE Access  
In this paper, we present a LU decomposition-based block-recursive algorithm for large-scale matrix inversion.  ...  Matrix inversion is a fundamental operation for solving linear equations for many computational applications, especially for various emerging big data applications.  ...  The implementation shows good performance, but it is not suitable for clusters. Yang, et al. [15] proposed a parallel algorithm for matrix inversion based on Gauss-Jordan elimination with pivoting.  ... 
doi:10.1109/access.2016.2546544 fatcat:npwy2xc4y5dqlbji2o4g2wniye

Comparison Criticality in Sorting Algorithms

Thomas B. Jones, David H. Ackley
2014 2014 44th Annual IEEE/IFIP International Conference on Dependable Systems and Networks  
Fault tolerance techniques often presume that the end-user computation must complete flawlessly.  ...  In a case study on traditional sorting algorithms, we present explorations of a criticality measure defined over expected fault damage rather than probability of correctness.  ...  core using fault tolerance techniques based in hardware, such as [5] - [8] , or in software, such as [9] - [13] .  ... 
doi:10.1109/dsn.2014.74 dblp:conf/dsn/JonesA14 fatcat:fftswzrvlne5dnjtxwaafqy4km

Numerical Defect Correction as an Algorithm-Based Fault Tolerance Technique for Iterative Solvers

Fabian Oboril, Mehdi B. Tahoori, Vincent Heuveline, Dimitar Lukarski, Jan-Philipp Weiss
2011 2011 IEEE 17th Pacific Rim International Symposium on Dependable Computing  
In this paper, we present an algorithm-based fault tolerant (ABFT) scheme for an iterative linear solver engine based on the Conjugated Gradient method (CG) by taking the advantage of numerical defect  ...  As hardware devices like processor cores and memory sub-systems based on nano-scale technologies nodes become more unreliable, the need for fault tolerant numerical computing engines, as used in many critical  ...  ACKNOWLEDGEMENTS The Shared Research Group 16-1 received financial support by the Concept for the Future of Karlsruhe Institute of Technology in the framework of the German Excellence Initiative and by  ... 
doi:10.1109/prdc.2011.26 dblp:conf/prdc/OborilTHLW11 fatcat:liad5fwnejek5arrqpbjbvb4qq

Real-number codes for fault-tolerant matrix operations on processor arrays

V.S.S. Nair, J.A. Abraham
1990 IEEE transactions on computers  
Abstmct-Various checksum codes have been suggested for fault-tolerant matrix computations on processor arrays.  ...  We prove that linearity is a necessary and sufficient condition for codes used for fault-tolerant matrix operations such as matrix addition, multiplication, transposition, and LU decomposition.  ...  The technique called algorithm-based fault tolerance (ABFT) has been suggested [6] for attaining the above objectives.  ... 
doi:10.1109/12.54836 fatcat:yajh3yoizvgutddkgavbuhq364

An efficient algorithm-based fault tolerance design using the weighted data-check relationship

Hee Yong Youn, Choong Gun Oh, Hyunseung Choo, Jin-Wook Chung, Dongman Lee
2001 IEEE transactions on computers  
Algorithm-based fault tolerance designs employing various encoding/decoding schemes have been proposed for such systems to effectively tolerate operation time fault.  ...  AbstractÐVLSI-based processor arrays have been widely used for computation intensive applications such as matrix and graph algorithms.  ...  To effectively tolerate the operation time fault in parallel systems where the number of faults is expected to be relatively small, Algorithm-Based Fault Tolerance (ABFT) designs have been developed [  ... 
doi:10.1109/12.919281 fatcat:ofmmaxj4dreatiyzlq3alsnmnq

Fast quantum subroutines for the simplex method [article]

Giacomo Nannicini
2022 arXiv   pre-print
(1) + d_c n) time in the worst case using the fastest known algorithm for sparse matrix multiplication.  ...  For example, for a problem with m constraints, n variables, at most d_c nonzero elements per column of the costraint matrix, at most d nonzero elements per column or row of the basis, basis condition number  ...  Acknowledgments We are grateful to Sergey Bravyi, Sanjeeb Dash, Santanu Dey, Yuri Faenza, Krzysztof Onak, Ted Yoder, and to anonymous referees for useful discussions and/or comments on an early version  ... 
arXiv:1910.10649v8 fatcat:lzgihxcarnbg3bl5y4tckoapym

Space-Efficient Fault-Tolerant Diameter Oracles [article]

Davide Bilò, Sarel Cohen, Tobias Friedrich, Martin Schirneck
2021 arXiv   pre-print
We design f-edge fault-tolerant diameter oracles (f-FDOs).  ...  The preprocessing time nearly matches a conditional lower bound for combinatorial algorithms, also by Henzinger et al.  ...  In this section, we assume the base graph G to be directed and present an (1 + ε)-approximate fault-tolerant diameter oracle with space O(m) and O(1) query time.  ... 
arXiv:2107.03485v1 fatcat:3a2l7rvbifdyzjiu4q23fhf7uu

A steepest edge active set algorithm for solving sparse linear programming problems

S. W. Sloan
1988 International Journal for Numerical Methods in Engineering  
The active set algorithm has a very simple geometric interpretation, works with an active constraint matrix of dimension n x n and is ideally suited to problems where n < m + r .  ...  A steepest edge active set algorithm is described which is suitable for solving linear programming problems where the constraint matrix is sparse and has more rows than columns.  ...  Kuczera for helpful discussions on linear programming strategies. He would also like to thank Ms Cherie Hook for typing the manuscript.  ... 
doi:10.1002/nme.1620261207 fatcat:crbtrnvpwjgvhc756qq46ftxxe

A Two-Step Screening Algorithm to Solve Linear Error Equations for Blind Identification of Block Codes Based on Binary Galois Field

2021 KSII Transactions on Internet and Information Systems  
The simulation results have shown that it can highly improve the fault tolerance ability compared to the existing Gauss elimination process-based algorithms.  ...  However, the fault tolerance will fall short when the intercepted bit error rate (BER) is too high.  ...  In this case, we must look for another method with strong fault tolerance to reconstruct the paritycheck matrix.  ... 
doi:10.3837/tiis.2021.09.020 fatcat:2nacipwowvhbjfzbdo5vb2taxu
« Previous Showing results 1 — 15 out of 1,098 results