Chapter 22 – Reconfigurable Computing

Publisher Summary The launch of static random access memory-based (SRAM) field-programmable gate arrays (FPGA) presented a new capability to the electronics fraternity: dynamically reconfigurable logic, which refers to designs that can be reconfigured on the fly while remaining resident in the system. SRAM-based variants allow the main system to download new configuration data into the device. Although all of the logic gates, registers, and SRAM cells forming the FPGA are created on the surface of a single piece of silicon substrate, it is sometimes useful to visualize the device as comprising two distinct strata: the logic gates/registers and the programmable SRAM configuration cells. Though it is great to be able to reconfigure the function of the individual devices on a circuit board, there are occasions when design engineers would like to create board-level systems that can be reconfigured to perform a variety of radically different functions. The solution to configure the board-level connections between devices are offered by field-programmable interconnect devices (FPIDs).