The Community for Technology Leaders
Field-Programmable Custom Computing Machines, Annual IEEE Symposium on (2006)
Napa, California
Apr. 24, 2006 to Apr. 26, 2006
ISBN: 0-7695-2661-6
pp: 205-216
Michael J. Wilson , California Institute of Technology
Michael Wrighton , California Institute of Technology
Henry Barnor , California Institute of Technology
Michael deLorimier , California Institute of Technology
Andre DeHon , California Institute of Technology
Raphael Rubin , California Institute of Technology
Nachiket Kapre , California Institute of Technology
Nikil Mehta , California Institute of Technology
Dedicated, spatially configured FPGA interconnect is efficient for applications that require high throughput connections between processing elements (PEs) but with a limited degree of PE interconnectivity (e.g. wiring up gates and datapaths). Applications which virtualize PEs may require a large number of distinct PE-to-PE connections (e.g. using one PE to simulate 100s of operators, each requiring input data from thousands of other operators), but with each connection having low throughput compared with the PE?s operating cycle time. In these highly interconnected conditions, dedicating spatial interconnect resources for all possible connections is costly and inefficient. Alternatively, we can time share physical network resources by virtualizing interconnect links, either by statically scheduling the sharing of resources prior to runtime or by dynamically negotiating resources at runtime. We explore the tradeoffs (e.g. area, route latency, route quality) between time-multiplexed and packetswitched networks overlayed on top of commodity FPGAs. We demonstrate modular and scalable networks which operate on a Xilinx XC2V6000-4 at 166MHz. For our applications, timemultiplexed, offline scheduling offers up to a 63% performance increase over online, packet-switched scheduling for equivalent topologies. When applying designs to equivalent area, packetswitching is up to 2? faster for small area designs while timemultiplexing is up to 5? faster for larger area designs. When limited to the capacity of a XC2V6000, if all communication is known, time-multiplexed routing outperforms packet-switching; however when the active set of links drops below 40% of the potential links, packet-switched routing can outperform timemultiplexing.
Michael J. Wilson, Michael Wrighton, Henry Barnor, Michael deLorimier, Andre DeHon, Raphael Rubin, Nachiket Kapre, Nikil Mehta, "Packet Switched vs. Time Multiplexed FPGA Overlay Networks", Field-Programmable Custom Computing Machines, Annual IEEE Symposium on, vol. 00, no. , pp. 205-216, 2006, doi:10.1109/FCCM.2006.55
99 ms
(Ver 3.3 (11022016))