期刊
2022 21ST INTERNATIONAL CONFERENCE ON FIELD-PROGRAMMABLE TECHNOLOGY (ICFPT 2022)
卷 -, 期 -, 页码 308-309出版社
IEEE
关键词
graph processing; memory semantic fabric
类别
资金
- Strategic Priority Research Program (Class A) of Chinese Academy of Sciences entitled Fundamental Software Ecosystem of RISC-V
- Beijing Municipal Education Commision [201914430004]
- National Natural Science Foundation of China [62090023, 62072439, 61702485]
- Youth Innovation Promotion Association of CAS [2017143]
- Beijing Municipal Natural Science Foundation [4212028]
- Shandong Provincial Natural Science Foundation [ZR2019LZH004]
FPGA has shown great potential for graph processing, but the limited memory capacity of a single FPGA can be a challenge. In this paper, we propose GraFF, a graph processing system that connects multiple FPGAs via a custom memory semantic Fabric. By optimizing system parallelism and relaxing FPGA synchronization, GraFF achieves significant performance improvement and exhibits linear scalability as the number of FPGAs increases.
FPGA has been a promising solution for graph processing in many scenarios. With a rapid growth in graph size, the on/off-chip memory capacity of a single FPGA is insufficient to hold large-scale graphs. To tackle such problem, in this position paper, we introduce GraFF, a Graph processing system with multiple FPGAs interconnected via a custom memory semantic Fabric. In order to efficiently exploit system parallelism, we first split the traversal of graph data into a series of independent fine-grained flits that are concurrently delivered among FPGAs as sheer memory semantic transactions. Then we relax FPGAs' synchronization from strict barrier boundaries between adjacent supersteps to fully parallelize graph traversing and computing. We build a prototype of GraFF with four custom FPGA nodes. Preliminary evaluation result based on the Breadth First Search (BFS) algorithm shows that the peak performance of GraFF reaches up to 6.23 GTEPS. Moreover, GraFF exhibits linear scalability when the number of FPGAs rises from one to four.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据