Abstract-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.
Abstract-Multiple techniques have now been proposed using random addressing to build demultiplexers which interface between the large pitch of lithographically patterned features and the smaller pitch of self-assembled sublithographic nanowires. At the same time, the relatively high defect rates expected for molecular-sized devices and wires dictate that we design architectures with spare components so we can map around defective elements. To accommodate and mask both of these effects, we introduce a programmable addressing scheme which can be used to provide deterministic addresses for decoders built with random nanoscale addressing and potentially defective wires. We describe how this programmable addressing scheme can be implemented with emerging, nanoscale building blocks and show how to build deterministically addressable memory banks. We characterize the area required for this programmable addressing scheme. For 2048 2048 memory banks, the area overhead for address correction is less than 33%, delivering net memory densities around 10 11 b/cm 2 .
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.