Fast Parallel Tensor Times Same Vector for Hypergraphs

Hypergraphs are a popular paradigm to represent complex real-world networks exhibiting multi-way relationships of varying sizes. Mining centrality in hyper-graphs via symmetric adjacency tensors has only recently become computationally feasible for large and complex datasets. To enable scalable comp...

Full description

Saved in:
Bibliographic Details
Main Authors: Shivakumar, Shruti, Amburg, Ilya, Aksoy, Sinan G., Li, Jiajia, Young, Stephen J., Aluru, Srinivas
Format: Conference Proceeding
Language:English
Subjects:
generating function
High performance computing
hypergraphs
Indexes
Instruction sets
Prediction algorithms
Scalability
sparse symmetric tensor times same vector
tensor eigenvector
Tensors
Vectors
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Hypergraphs are a popular paradigm to represent complex real-world networks exhibiting multi-way relationships of varying sizes. Mining centrality in hyper-graphs via symmetric adjacency tensors has only recently become computationally feasible for large and complex datasets. To enable scalable computation of these and related hypergraph analytics, here we focus on the Sparse Symmetric Tensor Times Same Vector (S 3 TTVC) operation. We introduce the Compound Compressed Sparse Symmetric (CCSS) format, an extension of the compact CSS format for hypergraphs of varying hyperedge sizes and present a shared-memory parallel algorithm to compute S 3 TTVC. We experimentally show S 3 TTVc computation using the CCSS format achieves better performance than the naive baseline, and is subsequently more performant for hypergraph H -eigenvector centrality.
ISSN:2640-0316
DOI:10.1109/HiPC58850.2023.00049