Abstract: A parallel processing system computer which utilizes the logic programming language Prolog comprising a plurality of processing nodes, each node comprising three central processing units (CPUs), a memory architecture adapted for Prolog execution and interfacing hardware, each processing node being connected to a communication bus and a real-time broadcast bus whereby the real-time data from an input can be broadcast via the real-time broadcast bus to each processing node. One of the CPUs is used to control the communications and scheduling of the node, the other two CPUs are used as sequential Prolog Processors (SPPS). Each node can be arranged such that the collection of unused memory is carried out by one SPP while another SPP continues to run Prolog program enabling continuous real-time operation. The memory architecture is hybrid and comprises local static RAM and global dynamic RAM. The dynamic RAM comprises a Prolog database of known signal for comparison with the real-time input signals.

Abstract: A multiprocessor in which a plurality of processors are connected via a system bus. The multiprocessor includes processor groups each having at least one processor, a main storage memory, a cache memory, a cache control unit, a directory memory, and a directory control unit. The directory control unit includes an invalidation command issuing unit that issues a cache line invalidation command to all reference designations stored in the directory memory in cache lines. Thus the device process time can be shortened by reducing the frequency that cache invalidation commands are issued from the cache memory.

Abstract: An apparatus for processing data has a Single-Instruction-Multiple-Data (SIMD) architecture, and a number of features that improve performance and programmability. The apparatus includes a rectangular array of processing elements and a controller. The apparatus offers a number of techniques for shifting image data within the array. A first technique, the ROLL option, simultaneously shifts image planes in opposite directions within the array. A second technique, the gated shift option, makes a normal shift of an image plane to neighboring PEs conditional, for each PE, upon a value stored in a mask register of each PE. A third technique, the carry propagate option, combines the computations from multiple PEs in order to complete an n-bit operation in fewer than n clocks by forming "supercells" within the array. The apparatus also includes a multi-bit X Pattern register and a multi-bit Y Pattern register.

Abstract: A distributed memory computer architecture associates separate memory blocks with their own processors, each of which executes the same program. A processor fetching data or instructions from its local memory also broadcasts that fetched data or instruction to the other processors to cut the time required for them to request this data. Runs of instruction and data local to one processor providing improved performance that is captured by the system as a whole by the ability of the other processors not executing local data or instructions to execute instructions out of order and return to find the data ready in buffer for rapid use.

Abstract: A byte accessible memory interface circuit using a reduced set of memory control signals. The present invention includes an interface circuit having a reduced set of memory control signals for performing word length reads and writes to an external memory module containing a plurality of integrated circuit (IC) memory chips. The interface circuit contains a respective multiplexer and a respective register circuit for each byte of the word length data. The multiplexers select a byte of data from either an on-chip data bus or from a bus carrying data read from the external memory module. To perform a full length word write, the data from the on-chip bus is loaded into the registers (via the multiplexers) and then written to the memory module. To perform a partial length word write, a pre-read operation is performed at the target address and a word length data is loaded into the registers. The new data is then received over the on-chip data bus and routed by the multiplexers into the byte locations to be changed.

Abstract: A NULL convention asynchronous register regulates wavefronts of signals through a NULL convention sequential circuit. The asynchronous register receives a control signal from a downstream circuit when the downstream circuit is ready to receive a new wavefront. Wavefronts alternate between meaningful data and NULL. The asynchronous register further generates a feedback signal to upstream circuits so that upstream circuits generate a wavefront when the sequential circuit is ready to receive the wavefront. Asynchronous registers can be used to create a variety of architectures, and to store data, as in a sequential circuit (state machine).

Abstract: The subject matter of the application essentially relates to a matrix array of processor units, each processor unit having, in addition to an arithmetic logic unit and a result register bank, a further arithmetic logic unit, a multiplier/adder unit, a storage unit of a distributed screen section buffer and a local general purpose memory. The processor is distinguished by a high processing speed in conjunction with a small chip area and enables real-time processing even in the case of computation-intensive image processing methods such as 2D convolution, Gabor transformation, Gaussian or Laplacian pyramids, block matching, DCT or MPEG2.

Abstract: A dynamically reconfigurable FPGA includes an array of tiles on a logic plane and a plurality of memory planes. Each tile has associated storage elements on each memory plane, called local memory. This local memory allows large amounts of data to pass from one FPGA configuration (memory plane) to another with no external memory access, thereby transferring data to/from the storage elements in the logic plane at very high speed. Typically, all the local memory can be simultaneously transferred to/from other memory planes in one cycle. Each FPGA configuration provides a virtual instruction. The present invention uses two different types of virtual instructions: computational and pattern manipulation instructions. Computational instructions perform some computation with data stored in some pre-defined local memory pattern. Pattern manipulation instructions move the local data into different memory locations to create the pattern required by the next instruction.

Abstract: A single-instruction multiple-data (SIMD) processor (10) that incorporates features for horizontal scaling of video data. The processor (10) has a data input register (11) that is operable to store input data word in sequential locations in the data input register (11) and transfer the input data words to an array of processing elements. The processor (10) also has an output data register (16) operable to receive data output words from the array of processing elements and to output said data output words from sequential locations of said output data array. An input skip signal input to the processor causes a sequential data write operation to skip a location of the input data register while an output skip signal to the processor causes a sequential data read operation to skip a location of the output data register.

Abstract: A routable operand and selectable operation processor multimedia extension unit is employed to motion compensate MPEG video using improved vector processing. A vector processing unit executes an add and divide instruction that adds two vector registers and divides the result in a single instruction. This is implemented through loading a first vector register with a first plurality of elements from a source block. A second vector register is then loaded with a second plurality of elements that are adjacent to the first plurality of elements. The add and divide instruction is then executed on the first and second vector registers, yielding an interpolated source element that is stored in a resultant vector register.

Abstract: A system and method for generating a program to enable reassignment of data items among processors in a massively-parallel computer to effect a predetermined rearrangement of address bits. The computer has a plurality of processing elements, each including a memory. Each memory includes a plurality of storage locations for storing a data item, each storage location within the computer being identified by an address, comprising a plurality of address bits having a global portion comprising a processing element identification portion and a local portion identifying the storage location within the memory of the particular processing element. The system generates a program to facilitate use of a predetermined set of tools to effect a reassignment of data items among processing elements and storage location to, in turn, effect a predetermined rearrangement of address bits. The system includes a global processing portion and a local processing portion.

Abstract: A data processing system having flexibility coping with parallelism of a program comprises a plurality of processor elements for executing instructions, a main memory shared by the plurality of processor elements, and a plurality of parallel operation control facilities for enabling the plurality of processor elements to operate in synchronism. The plurality of parallel operation control facilities are provided in correspondence to the plurality of processor elements, respectively. The data processing system further comprises a multiprocessor operation control facility for enabling the plurality of processor elements to operate independently, and a flag for holding a value indicating which of the parallel operation mode or the multiprocessor mode is to be activated. The shared cache memory is implemented in a blank instruction and controlled by a cache controller so that inconsistency of the data stored in the cache memory is eliminated.