Graph neural networks (GNNs) process large-scale graphs consisting of a hundred billion edges. In contrast to traditional deep learning, unique behaviors of the emerging GNNs are engaged with a large set of graphs and embedding data on storage, which exhibits complex and irregular preprocessing. We propose a novel deep learning framework on large graphs, HolisticGNN, that provides an easy-to-use, near-storage inference infrastructure for fast, energy-efficient GNN processing. To achieve the best end-to-end latency and high energy efficiency, HolisticGNN allows users to implement various GNN algorithms and directly executes them where the actual data exist in a holistic manner. It also enables RPC over PCIe such that the users can simply program GNNs through a graph semantic library without any knowledge of the underlying hardware or storage configurations. We fabricate HolisticGNN's hardware RTL and implement its software on an FPGA-based computational SSD (CSSD). Our empirical evaluations show that the inference time of HolisticGNN outperforms GNN inference services using high-performance modern GPUs by 7.1× while reducing energy consumption by 33.2×, on average.
|Title of host publication||Proceedings of the 20th USENIX Conference on File and Storage Technologies, FAST 2022|
|Number of pages||17|
|Publication status||Published - 2022|
|Event||20th USENIX Conference on File and Storage Technologies, FAST 2022 - Santa Clara, United States|
Duration: 2022 Feb 22 → 2022 Feb 24
|Name||Proceedings of the 20th USENIX Conference on File and Storage Technologies, FAST 2022|
|Conference||20th USENIX Conference on File and Storage Technologies, FAST 2022|
|Period||22/2/22 → 22/2/24|
Bibliographical noteFunding Information:
This research is supported by Samsung Research Funding & Incubation Center of Samsung Electronics (SRFC-IT2101-04). This work is protected by one or more patents, and Myoungsoo Jung is the corresponding author. The authors would like to thank the anonymous reviewers for their comments and suggestions. The authors also thank Raju Rangaswami for shepherding this paper.
© AST 2022.All rights reserved.
All Science Journal Classification (ASJC) codes
- Computer Networks and Communications
- Hardware and Architecture