Dr. Top-k: Delegate-Centric Top-k on GPUs

TL;DR

Dr. Top-k introduces a delegate-centric GPU system that significantly reduces top-$k$ workloads, employs theoretical analysis for optimal subrange sizing, and enables efficient multi-GPU top-$k$ computation, outperforming existing methods.

Contribution

The paper presents a novel delegate-centric design, theoretical analysis for subrange sizing, and four system optimizations for multi-GPU top-$k$ processing.

Findings

Reduces top-$k$ workload by over 99%

Outperforms state-of-the-art methods

Enables fast multi-GPU top-$k$ computation

Abstract

Recent top-$k$ computation efforts explore the possibility of revising various sorting algorithms to answer top-$k$ queries on GPUs. These endeavors, unfortunately, perform significantly more work than needed. This paper introduces Dr. Top-k, a Delegate-centric top-$k$ system on GPUs that can reduce the top-$k$ workloads significantly. Particularly, it contains three major contributions: First, we introduce a comprehensive design of the delegate-centric concept, including maximum delegate, delegate-based filtering, and $\beta$ delegate mechanisms to help reduce the workload for top-$k$ up to more than 99%. Second, due to the difficulty and importance of deriving a proper subrange size, we perform a rigorous theoretical analysis, coupled with thorough experimental validations to identify the desirable subrange size. Third, we introduce four key system optimizations to enable fast multi-GPU top-$k$ computation. Taken together, this work constantly outperforms the state-of-the-art.