Optimizing Datalog for the GPU

TL;DR

This paper introduces GPUlog, a GPU-optimized Datalog engine that significantly accelerates recursive query processing, achieving up to 45x speedups over CPU-based engines in static analysis tasks.

Contribution

The paper presents GPUlog, a novel GPU-based Datalog engine utilizing the HISA data structure for efficient in-memory recursive query execution.

Findings

GPUlog achieves up to 45x speedup over Soufflé.

HISA enables efficient range-based operations on dense data.

GPUlog has a favorable memory footprint compared to GPU-based joins.

Abstract

Modern Datalog engines (e.g., LogicBlox, Souffl\'e, ddlog) enable their users to write declarative queries which compute recursive deductions over extensional facts, leaving high-performance operationalization (query planning, semi-na\"ive evaluation, and parallelization) to the engine. Such engines form the backbone of modern high-throughput applications in static analysis, network monitoring, and social-media mining. In this paper, we present a methodology for implementing a modern in-memory Datalog engine on data center GPUs, allowing us to achieve significant (up to 45x) gains compared to Souffl\'e (a modern CPU-based engine) on context-sensitive points-to analysis of httpd. We present GPUlog, a Datalog engine backend that implements iterated relational algebra kernels over a novel range-indexed data structure we call the hash-indexed sorted array (HISA). HISA combines the algorithmic benefits of incremental range-indexed relations with the raw computation throughput of operations over dense data structures. Our experiments show that GPUlog is significantly faster than CPU-based Datalog engines while achieving a favorable memory footprint compared to contemporary GPU-based joins.