Towards Programmable Memory Controller for Tensor Decomposition

Sasindu Wijeratne, Ta-Yang Wang, Rajgopal Kannan, Viktor Prasanna

2022

Abstract

Tensor decomposition has become an essential tool in many data science applications. Sparse Matricized Tensor Times Khatri-Rao Product (MTTKRP) is the pivotal kernel in tensor decomposition algorithms that decompose higher-order real-world large tensors into multiple matrices. Accelerating MTTKRP can speed up the tensor decomposition process immensely. Sparse MTTKRP is a challenging kernel to accelerate due to its irregular memory access characteristics. Implementing accelerators on Field Programmable Gate Array (FPGA) for kernels such as MTTKRP is attractive due to the energy efficiency and the inherent parallelism of FPGA. This paper explores the opportunities, key challenges, and an approach for designing a custom memory controller on FPGA for MTTKRP while exploring the parameter space of such a custom memory controller.

Paper Citation

in Harvard Style

Wijeratne S., Wang T., Kannan R. and Prasanna V. (2022). Towards Programmable Memory Controller for Tensor Decomposition. In Proceedings of the 11th International Conference on Data Science, Technology and Applications - Volume 1: DATA, ISBN 978-989-758-583-8, pages 468-475. DOI: 10.5220/0011301200003269

in Bibtex Style

@conference{data22,
author={Sasindu Wijeratne and Ta-Yang Wang and Rajgopal Kannan and Viktor Prasanna},
title={Towards Programmable Memory Controller for Tensor Decomposition},
booktitle={Proceedings of the 11th International Conference on Data Science, Technology and Applications - Volume 1: DATA,},
year={2022},
pages={468-475},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011301200003269},
isbn={978-989-758-583-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 11th International Conference on Data Science, Technology and Applications - Volume 1: DATA,
TI - Towards Programmable Memory Controller for Tensor Decomposition
SN - 978-989-758-583-8
AU - Wijeratne S.
AU - Wang T.
AU - Kannan R.
AU - Prasanna V.
PY - 2022
SP - 468
EP - 475
DO - 10.5220/0011301200003269