Exploring irregular memory accesses on FPGAs

Authors:
Robert J. Halstead;Jason Villarreal;Walid Najjar
Affiliations:
University of California, Riverside, Riverside, CA, USA;Jacquard Computing, Riverside, CA, USA;University of California, Riverside, Riverside, CA, USA
Venue:
Proceedings of the first workshop on Irregular applications: architectures and algorithm
Year:
2011

Citing 9
Cited 1

The horizon supercomputing system: architecture and software

Proceedings of the 1988 ACM/IEEE conference on Supercomputing
A processor architecture for horizon

Proceedings of the 1988 ACM/IEEE conference on Supercomputing
Exploiting heterogeneous parallelism on a multithreaded multiprocessor

ICS '92 Proceedings of the 6th international conference on Supercomputing
The Tera computer system

ICS '90 Proceedings of the 4th international conference on Supercomputing
Multi-processor performance on the Tera MTA

SC '98 Proceedings of the 1998 ACM/IEEE conference on Supercomputing
ELDORADO

Proceedings of the 2nd conference on Computing frontiers
Early Experiences on the NRL Cray XD1

HPCMP-UGC '06 Proceedings of the HPCMP Users Group Conference
Instruction Set Innovations for the Convey HC-1 Computer

IEEE Micro
Designing Modular Hardware Accelerators in C with ROCCC 2.0

FCCM '10 Proceedings of the 2010 18th IEEE Annual International Symposium on Field-Programmable Custom Computing Machines

Compiled multithreaded data paths on FPGAs for dynamic workloads

Proceedings of the 2013 International Conference on Compilers, Architectures and Synthesis for Embedded Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Algorithms that exhibit irregular memory access patterns are known to show poor performance on multiprocessor architectures, particularly when memory access latency is variable. Many common data structures, including graphs, trees, and linked-lists, exhibit these irregular memory access patterns. While FPGA-based code accelerators have been successful on applications with regular memory access patterns, they have not been further explored for irregular memory access patterns. Multithreading has been shown to be an effective technique in masking long latencies. We describe the compiler generation of concurrent hardware threads for FPGAs with the objective of masking the memory latency caused by irregular memory access patterns. We extend the ROCCC compiler to generate customized state information for each dynamically generated thread.