GraphStep: A System Architecture for Sparse-Graph Algorithms
Abstract
Many important applications are organized around long-lived, irregular sparse graphs (e.g., data and knowledge bases, CAD optimization, numerical problems, simulations). The graph structures are large, and the applications need regular access to a large, data-dependent portion of the graph for each operation (e.g., the algorithm may need to walk the graph, visiting all nodes, or propagate changes through many nodes in the graph). On conventional microprocessors, the graph structures exceed on-chip cache capacities, making main-memory bandwidth and latency the key performance limiters. To avoid this "memory wall," we introduce a concurrent system architecture for sparse graph algorithms that places graph nodes in small distributed memories paired with specialized graph processing nodes interconnected by a lightweight network. This gives us a scalable way to map these applications so that they can exploit the high-bandwidth and low-latency capabilities of embedded memories (e.g., FPGA Block RAMs). On typical spreading activation queries on the ConceptNet Knowledge Base, a sample application, this translates into an order of magnitude speedup per FPGA compared to a state-of-the-art Pentium processor.
Additional Information
© 2006 IEEE. Issue Date: 24-26 April 2006, Date of Current Version: 11 December 2006. This work was supported in part by DARPA under grant FA8750-05-C-0011, the NSF CAREER program under grant CCR-0133102, and the Microelectronics Advanced Research Consortium (MARCO) as part of the efforts of the Gigascale Systems Research Center (GSRC). Xilinx Corporation donated hardware, including the XC2V6000s used for the FPGA implementation.Attached Files
Files
Name | Size | Download all |
---|---|---|
md5:52259ace53e41b0dadc668e09307d7e5
|
181.2 kB | Preview Download |
Additional details
- Eprint ID
- 22460
- Resolver ID
- CaltechAUTHORS:20110223-132654232
- FA8750-05-C-0011
- Defense Advanced Research Projects Agency (DARPA)
- CCR-0133102
- NSF
- Microelectronics Advanced Research Consortium (MARCO)
- Gigascale Systems Research Center
- Created
-
2011-02-23Created from EPrint's datestamp field
- Updated
-
2021-11-09Created from EPrint's last_modified field
- Series Name
- Annual IEEE Symposium on Field-Programmable Custom Computing Machines