Abstract: A power-efficient, distributed reconfigurable computing system and method are provided. A reconfigurable computing system may include an embedded controller for performing real-time control and initialization and circuitry that supports data-flow driven execution of processing phases. The circuitry may include processing elements such as RAM-based field programmable gate array devices and direct memory access engines. The processing elements can be configured for one or more functions or operations of a program and then reconfigured for other functions or operations of the program. The processing elements can be configured or reconfigured to construct a desired sequence of operations in real-time. A processing element may be divided into slots, each of which includes a substantially similar amount of resources. A processing element includes one or more wrappers, and a wrapper may occupy a slot or a group of slots. Layered software architecture separates control software from implementation software.