Hedgehog: a Performance-Oriented General Purpose Library for Multi-GPU Systems - Archive ouverte HAL Access content directly
Conference Papers Year : 2020

Hedgehog: a Performance-Oriented General Purpose Library for Multi-GPU Systems

(1, 2) , (1) , (2) , (1) , (2)
1
2

Abstract

We present Hedgehog, a general-purpose library for taking advantage of powerful compute nodes, multicore CPUs, and multiple GPUs. The novel aspects of Hedgehog are: (1) its explicit representation of a program as a dataflow graph, (2) its pure dataflow-driven scheduling, (3) its maintenance of a computation's localized state via state managers, and (4) its fine control of memory via memory managers. This dataflow approach results in extremely low overhead for task executions (< 1 microsecond) and no-cost profiling at the task level. This allows us to prototype operations that compare favorably with leading libraries such as cuBLAS-XT.
Vignette du fichier
presentation.pdf (4.1 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02981031 , version 1 (01-04-2022)

Identifiers

  • HAL Id : hal-02981031 , version 1

Cite

Alexandre Bardakoff, Timothy Blattner, Bruno Bachelet, Walid Keyrouz, Loïc Yon. Hedgehog: a Performance-Oriented General Purpose Library for Multi-GPU Systems. GPU Technology Conference (GTC), NVIDIA, Mar 2020, San Diego, United States. ⟨hal-02981031⟩
89 View
0 Download

Share

Gmail Facebook Twitter LinkedIn More