AMPED: Accelerating MTTKRP for Billion-Scale Sparse Tensor Decomposition on Multiple GPUs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wijeratne, Sasindu, Kannan, Rajgopal, Prasanna, Viktor
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912529536516096
author Wijeratne, Sasindu
Kannan, Rajgopal
Prasanna, Viktor
author_facet Wijeratne, Sasindu
Kannan, Rajgopal
Prasanna, Viktor
contents Matricized Tensor Times Khatri-Rao Product (MTTKRP) is the computational bottleneck in sparse tensor decomposition. As real-world sparse tensors grow to billions of nonzeros, they increasingly demand higher memory capacity and compute throughput from hardware accelerators. In this work, we present AMPED, a multi-GPU parallel algorithm designed to accelerate MTTKRP on billion-scale sparse tensors. AMPED scales beyond the limits of a single GPU, meeting both the memory and performance requirements of large-scale workloads. We introduce a partitioning strategy combined with a dynamic load balancing scheme to distribute computation and minimize GPU idle time. On real-world billion-scale tensors, AMPED achieves a 5.1x geometric mean speedup in total execution time over state-of-the-art GPU baselines using 4 GPUs on a single CPU node.
format Preprint
id arxiv_https___arxiv_org_abs_2507_15121
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AMPED: Accelerating MTTKRP for Billion-Scale Sparse Tensor Decomposition on Multiple GPUs
Wijeratne, Sasindu
Kannan, Rajgopal
Prasanna, Viktor
Distributed, Parallel, and Cluster Computing
Matricized Tensor Times Khatri-Rao Product (MTTKRP) is the computational bottleneck in sparse tensor decomposition. As real-world sparse tensors grow to billions of nonzeros, they increasingly demand higher memory capacity and compute throughput from hardware accelerators. In this work, we present AMPED, a multi-GPU parallel algorithm designed to accelerate MTTKRP on billion-scale sparse tensors. AMPED scales beyond the limits of a single GPU, meeting both the memory and performance requirements of large-scale workloads. We introduce a partitioning strategy combined with a dynamic load balancing scheme to distribute computation and minimize GPU idle time. On real-world billion-scale tensors, AMPED achieves a 5.1x geometric mean speedup in total execution time over state-of-the-art GPU baselines using 4 GPUs on a single CPU node.
title AMPED: Accelerating MTTKRP for Billion-Scale Sparse Tensor Decomposition on Multiple GPUs
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2507.15121