TCUDB: Accelerating Database with Tensor Processors

TL;DR

TCUDB leverages NVIDIA's Tensor Core Units to accelerate database query processing, achieving up to 288x speedup in real-world applications by using matrix operations for core query operators.

Contribution

This work introduces TCUDB, a novel database engine that utilizes tensor processors for accelerating query operators, a strategy largely unexplored in traditional GPU-based databases.

Findings

Up to 288x speedup over baseline GPU query engine.

Effective acceleration of natural joins and group-by aggregates.

Significant performance improvements in entity matching, graph queries, and data analytics.

Abstract

The emergence of novel hardware accelerators has powered the tremendous growth of machine learning in recent years. These accelerators deliver incomparable performance gains in processing high-volume matrix operators, particularly matrix multiplication, a core component of neural network training and inference. In this work, we explored opportunities of accelerating database systems using NVIDIA's Tensor Core Units (TCUs). We present TCUDB, a TCU-accelerated query engine processing a set of query operators including natural joins and group-by aggregates as matrix operators within TCUs. Matrix multiplication was considered inefficient in the past; however, this strategy has remained largely unexplored in conventional GPU-based databases, which primarily rely on vector or scalar processing. We demonstrate the significant performance gain of TCUDB in a range of real-world applications including entity matching, graph query processing, and matrix-based data analytics. TCUDB achieves up to 288x speedup compared to a baseline GPU-based query engine.