7,120 Hits in 3.4 sec

Minimizing Congestion for Balanced Dominators [article]

Yosuke Mizutani, Annie Staker, Blair D. Sullivan
2022 arXiv   pre-print
We complement our theoretical results with computational experiments on a corpus of real-world networks showing sparse dominating sets lead to more balanced neighborhood partitionings.  ...  This leads to balanced neighborhood partitioning: given an r-dominating set, find a partition into connected subgraphs with optimal uniformity so that each vertex is co-assigned with some closest dominator  ...  ACKNOWLEDGMENTS Special thanks to Rachel Walker for early discussions on notions of sparse dominating sets & related terminology, and to Titus Brown & Taylor Reiter for assistance with metagenomic datasets  ... 
arXiv:2112.10973v2 fatcat:cnzo2n2fsjg23i3aruzjcowmeu

Communication Sparsity in Distributed Spiking Neural Network Simulations to Improve Scalability

Carlos Fernandez-Musoles, Daniel Coca, Paul Richmond
2019 Frontiers in Neuroinformatics  
Communication is recognized as the dominant part of distributed SNN simulations.  ...  Results show a combined gain when using hypergraph-based allocation and dynamic sparse communication, increasing computational efficiency by up to 40.8 percentage points and reducing simulation time by  ...  PR and DC performed substantial critical revisions and contributed to the experimental design. CF-M implemented the code.  ... 
doi:10.3389/fninf.2019.00019 pmid:31001102 pmcid:PMC6454199 fatcat:nuiulri5ojhk5agtuiq2igai34

Sparse Surface Speed Evaluation on a Dynamic Three-Dimensional Surface Using an Iterative Partitioning Scheme [chapter]

Paul Manstetten, Lukas Gnam, Andreas Hössinger, Siegfried Selberherr, Josef Weinbub
2018 Lecture Notes in Computer Science  
This sparse set of points is generated according to application-specific requirements using an iterative partitioning scheme.  ...  We diffuse the result of a constant extrapolation in the neighborhood of the sparse points to obtain an approximation to a linear interpolation between the sparse points.  ...  The financial support by the Austrian Federal Ministry of Science, Research and Economy and the National Foundation for Research, Technology and Development is gratefully acknowledged.  ... 
doi:10.1007/978-3-319-93698-7_53 fatcat:k3uj5ztulrfrxd4fmqr7ua52mi

Domain decomposition and locality optimization for large-scale lattice Boltzmann simulations

M. Wittmann, T. Zeiser, G. Hager, G. Wellein
2013 Computers & Fluids  
We present a simple, parallel and distributed algorithm for setting up and partitioning a sparse representation of a regular discretized simulation domain.  ...  This method is scalable for a large number of processes even for complex geometries and ensures load balance between the domains, reasonable communication interfaces, and good data locality within the  ...  Lively discussions with Jan Treibig, and Johannes Habich are gratefully acknowledged. This work was supported by BMBF under grant No. 01IH08003A (project SKALB).  ... 
doi:10.1016/j.compfluid.2012.02.007 fatcat:w6uwycdvqbev7pdzr3spwz62si

ByteGNN: Efficient Graph Neural Network Training at Large Scale

Chenguang Zheng, Hongzhi Chen, Yuxuan Cheng, Zhezheng Song, Yifan Wu, Changji Li, James Cheng, Hao Yang, Shuai Zhang
2022 Proceedings of the VLDB Endowment  
to reduce the end-to-end GNN training time, and (3) a graph partitioning algorithm tailored for GNN workloads.  ...  Graph neural networks (GNNs) have shown excellent performance in a wide range of applications such as recommendation, risk control, and drug discovery.  ...  The work of Chenguang Zheng, Changji Li and James Cheng was partially supported by a ByteDance Research Collaboration Project.  ... 
dblp:journals/pvldb/ZhengCCSWLCYZ22 fatcat:ow7kqw7cz5c5xpfiy3n6trchl4

Parallel Multiprojection Preconditioned Methods Based on Subspace Compression

Byron E. Moutafis, Christos K. Filelis-Papadopoulos, George A. Gravvanis
2017 Mathematical Problems in Engineering  
During the last decades, the continuous expansion of supercomputing infrastructures necessitates the design of scalable and robust parallel numerical methods for solving large sparse linear systems.  ...  A new approach for the additive projection parallel preconditioned iterative method based on semiaggregation and a subspace compression technique, for general sparse linear systems, is presented.  ...  The neighborhood distance parameter has been set to one, unless stated otherwise explicitly.  ... 
doi:10.1155/2017/2580820 fatcat:lsugkfu5fffi5lox25wgz2jzvu

Image Transformation Based on Learning Dictionaries across Image Spaces

Kui Jia, Xiaogang Wang, Xiaoou Tang
2013 IEEE Transactions on Pattern Analysis and Machine Intelligence  
of corresponding source and target image patches have the same support, i.e., the same indices of nonzero elements. (2) We devise a space partitioning scheme to divide the high-dimensional but sparse  ...  After coupled dictionary learning, sparse coefficient vectors of training image patch pairs are partitioned into easily retrievable local clusters.  ...  The parameter settings of sparse coding and space partitioning in the source space were same for these three methods. We set ζ = 20 and η = 9.  ... 
doi:10.1109/tpami.2012.95 pmid:22529324 fatcat:b26jvrj4w5g3tpt2ayqpru4nru

A Systematic Survey of General Sparse Matrix-Matrix Multiplication [article]

Jianhua Gao, Weixing Ji, Zhaonian Tan, Yueyan Zhao
2020 arXiv   pre-print
Covered topics include SpGEMM applications, size prediction of result matrix, matrix partitioning and load balancing, result accumulating, and target architecture-oriented optimization.  ...  SpGEMM (General Sparse Matrix-Matrix Multiplication) has attracted much attention from researchers in fields of multigrid methods and graph analysis.  ...  Matrix Partition and Load Balancing The second problem is matrix partitioning and load balancing.  ... 
arXiv:2002.11273v1 fatcat:5ppccisodvaevdhvfhawvjam5q

More Compact Oracles for Approximate Distances in Undirected Planar Graphs [chapter]

Ken-ichi Kawarabayashi, Christian Sommer, Mikkel Thorup
2013 Proceedings of the Twenty-Fourth Annual ACM-SIAM Symposium on Discrete Algorithms  
The tradeoff between the space requirements and the query time of distance oracles is of particular interest and the main focus of this paper.  ...  Unless stated otherwise, we assume all graphs to be planar and undirected. In FOCS 2001 (J.  ...  Thanks also to various anonymous reviewers for their comments and suggestions.  ... 
doi:10.1137/1.9781611973105.40 dblp:conf/soda/KawarabayashiST13 fatcat:bw7b44ybkna23nk7sjyx3hxyw4

A Simple Algorithm for the Graph Minor Decomposition − Logic meets Structural Graph Theory– [chapter]

Martin Grohe, Ken-ichi Kawarabayashi, Bruce Reed
2013 Proceedings of the Twenty-Fourth Annual ACM-SIAM Symposium on Discrete Algorithms  
The tradeoff between the space requirements and the query time of distance oracles is of particular interest and the main focus of this paper.  ...  Unless stated otherwise, we assume all graphs to be planar and undirected. In FOCS 2001 (J.  ...  Thanks also to various anonymous reviewers for their comments and suggestions.  ... 
doi:10.1137/1.9781611973105.30 dblp:conf/soda/GroheKR13 fatcat:hzy72uuhk5c3vjliv2izydtbqe

Parallel algorithms for forward and back substitution in direct solution of sparse linear systems

Anshul Gupta, Vipin Kumar
1995 Proceedings of the 1995 ACM/IEEE conference on Supercomputing (CDROM) - Supercomputing '95  
Our analysis and experiments show that, although not as scalable as the best parallel sparse Cholesky factorization algorithms, parallel sparse triangular solvers can yield reasonable speedups in runtime  ...  A few parallel algorithms for solving triangular systems resulting from parallel factorization of sparse linear systems have been proposed and implemented recently.  ...  For a triangular system resulting from the factorization of an N × N sparse matrix corresponding to a three-dimensional neighborhood graph, W = O(N 4/3 ) (7) and T o = O( p 2 ) + O( pN 2/3 ). (8) Balancing  ... 
doi:10.1145/224170.224471 dblp:conf/sc/GuptaK95 fatcat:nfqpljy7wredncqddhmq44s2xm

Guaranteed-Delivery Geographic Routing Under Uncertain Node Locations

S. Funke, N. Milosavljevic
2007 IEEE INFOCOM 2007 - 26th IEEE International Conference on Computer Communications  
Due to its macroscopic view, our routing scheme also produces shorter and more load-balanced paths than common geographic routing schemes, in particular in sparsely connected networks or in the presence  ...  In this paper we propose a novel naming and routing scheme that can handle the uncertainty in location information.  ...  The idea is to approximate the Voronoi and the Delaunay diagram of a sparse set of landmark nodes (analogous to a set of points in the plane), using the rest of the nodes as a sampling of the plane.  ... 
doi:10.1109/infcom.2007.148 dblp:conf/infocom/FunkeM07 fatcat:t7hm5ykmgbfs3ndmwur2qocmwy

Gemini: A Computation-Centric Distributed Graph Processing System

Xiaowei Zhu, Wenguang Chen, Weimin Zheng, Xiaosong Ma
2016 USENIX Symposium on Operating Systems Design and Implementation  
, plus (5) locality-aware chunking and fine-grained work-stealing for improving both inter-node and intra-node load balance, respectively.  ...  Gemini adopts (1) a sparse-dense signal-slot abstraction to extend the hybrid push-pull computation model from shared-memory to distributed scenarios, (2) a chunk-based partitioning scheme enabling low-overhead  ...  Acknowledgments We sincerely thank all the reviewers for their insightful comments and suggestions.  ... 
dblp:conf/osdi/ZhuCZM16 fatcat:lplpvidq7bh2jbt2qjwdkq62jq

Two novel multiway circuit partitioning algorithms using relaxed locking

A. Dasdan, C. Aykanat
1997 IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems  
This paper also provides guidelines for good parameter settings for the generic algorithms.  ...  Each approach leads to KL-based generic algorithms whose parameters can be set to obtain algorithms with different performance characteristics.  ...  The cutset of a partition is the set of all cut nets. The multiway circuit partitioning problem involves a way partitioning of such that the cutsize is minimized and the partitioning is balanced.  ... 
doi:10.1109/43.573831 fatcat:ehjyec4d5zb6reybbkbfuyqc3i

Scalable communication protocols for dynamic sparse data exchange

Torsten Hoefler, Christian Siebert, Andrew Lumsdaine
2010 SIGPLAN notices  
We define the dynamic sparse data-exchange (DSDE) problem and derive bounds in the well known LogGP model.  ...  To enable the compact and efficient expression of the communication phase, we develop suitable sparse communication protocols for irregular applications at large scale.  ...  We thank Jesper Larsson Träff (NEC) and Rolf Rabenseifner (HLRS) for discussions about sparse communications and protocols, Peter Gottschling (TUD) for comments about sparse matrix-vector multiplications  ... 
doi:10.1145/1837853.1693476 fatcat:pkrs4kt7xvcdjpjgkifduktnc4
« Previous Showing results 1 — 15 out of 7,120 results