Gravel: Fine-Grain GPU-Initiated Network Messages | IEEE Conference Publication | IEEE Xplore

Gravel: Fine-Grain GPU-Initiated Network Messages


Abstract:

Distributed systems incorporate GPUs because they provide massive parallelism in an energy-efficient manner. Unfortunately, existing programming models make it difficult ...Show More

Abstract:

Distributed systems incorporate GPUs because they provide massive parallelism in an energy-efficient manner. Unfortunately, existing programming models make it difficult to route a GPU-initiated network message. The traditional coprocessor model forces programmers to manually route messages through the host CPU. Other models allow GPU-initiated communication, but are inefficient for small messages. To enable fine-grain PGAS-style communication between threads executing on different GPUs, we introduce Gravel. GPU-initiated messages are offloaded through a GPU-efficient concurrent queue to an aggregator (implemented with CPU threads), which combines messages targeting to the same destination. Gravel leverages diverged work-group-level semantics to amortize synchronization across the GPU’s data-parallel lanes. Using Gravel, we can distribute six applications, each with frequent small messages, across a cluster of eight GPU-accelerated nodes. Compared to one node, these applications run 5.3x faster, on average. Furthermore, we show Gravel is more programmable and usually performs better than prior GPU networking models. CCS CONCEPTS Computer methodologies→Massively parallel algorithms;
Date of Conference: 12-17 November 2017
Date Added to IEEE Xplore: 27 October 2022
Electronic ISBN:978-1-4503-5114-0

ISSN Information:

Conference Location: Denver, CO, USA

References

References is not available for this document.