Latency-Optimized Networks for Clustering FPGAs

The data-intensive applications that will shape computing in the coming decades require scalable architectures that incorporate scalable data and compute resources and can support random requests to unstructured (e.g., logs) and semi-structured (e.g., large graph and XML) data sets. To explore the suitability of FPGAs for these computations, the authors are constructing an FPGA-based system with a memory capacity of 512 GB from a collection of 32 Virtex-5 FPGAs spread across 8 enclosures. This paper describes their work in exploring alternative interconnect technologies and network topologies for FPGA-based clusters.

Provided by: University of Calgary Topic: Hardware Date Added: Apr 2013 Format: PDF

Find By Topic