Wasp: Efficient Asynchronous Single-Source Shortest Path on Multicore Systems via Work Stealing
Paper i proceeding, 2025

The Single-Source Shortest Path (SSSP) problem is a fundamental graph problem with an extensive set of real-world applications. State-of-the-art parallel algorithms for SSSP, such as the Δ-stepping algorithm, create parallelism through priority coarsening. Priority coarsening results in redundant computations that diminish the benefits of parallelization and limit parallel scalability. This paper introduces Wasp, a novel SSSP algorithm that reduces parallelism-induced redundant work by utilizing asynchrony and an efficient priority-aware work stealing scheme. Contrary to previous work, Wasp introduces redundant computations only when threads have no high-priority work locally available to execute. This is achieved by a novel priority-aware work stealing mechanism that controls the inefficiencies of indiscriminate priority coarsening. Experimental evaluation shows competitive or better performance compared to GAP, GBBS, MultiQueues, Galois, Δ-stepping, and ρ-stepping on 13 diverse graphs with geometric mean speedups of 2.23× on AMD Zen 3 and 2.16× on Intel Sapphire Rapids using 128 threads.

Graph Algorithms

Shared-Memory

Single-Source Shortest Path

Författare

Marco D'antonio

Queen's University Belfast

Thai Son Mai

Queen's University Belfast

Philippas Tsigas

Chalmers, Data- och informationsteknik, Dator- och nätverkssystem

Göteborgs universitet

Hans Vandierendonck

Queen's University Belfast

Proceedings of the International Conference for High Performance Computing Networking Storage and Analysis Sc 2025

2109-2125
9798400714665 (ISBN)

2025 International Conference for High Performance Computing, Networking, Storage, and Analysis, SC 2025
St. Louis, USA,

Relaxed Semantics Across the Data Analytics Stack (RELAX-DN)

Europeiska kommissionen (EU) (EC/HE/101072456), 2023-03-01 -- 2027-03-01.

Ämneskategorier (SSIF 2025)

Datavetenskap (datalogi)

DOI

10.1145/3712285.3759872

Mer information

Senast uppdaterat

2025-12-23