SIMD Algorithms For 2 - D Arrays In Shuffle Networks
نویسندگان
چکیده
This paper studies a set of 2-D basic algorithms for SIMD perfect shuffle networks. Those algorithms where studied in [1] and [SN], but for the 1-D case, where the size of the problem N is the same as the number of processors P. For the 2-D case of N = L * P, we improve some algorithms to be "2-D efficient", by improving their running time from Ο(L *logP) to Ο(L + logP), as N exceeds P. We give nontrivial algorithms for the following operations: Row-Reduction, Parallel-Prefix, Transpose, Packing, Smoothing and Cartesian-Product.
منابع مشابه
Performing the Shuffle with the PM2I and Illiac SIMD Interconnection Networks
Three SIMD single stage interconnection networks which have been proposed and studied in the literature are the Illiac, PM2I, and Shuffle-Exchange. Here the ability of the Illiac and PM2I networks to perform the shuffle interconnection in an SIMD machine with N processors is examined. A lower bound of 3\/N/2 transfers for the Illiac to shuffle data is derived. An algorithm to do this task in 2\...
متن کاملRecirculating , Pipelined , and Multistage Simd Interconnection Networks
Hardware implementations of the Cube, Shuffle-Exchange, and Plus-Mi us-2 (PM2I) networks are examined. An analysis is made of recirculating (single stage) networks, multistage networks of combinational logic, and pipelined multistage networks. An expanded control structure for recirculating networks is introduced. It allows each processor to select its interconnection function independently of ...
متن کامل2 Fractal Image Compression - the Encoding Phase 2
In this work we introduce and analyze algorithms for fractal image compression on massively parallel SIMD arrays. The diierent algorithms discussed diier signiicantly in terms of their communication and computation structure. Therefore the most suited algorithm for a given architecture may be selected according to our investigations. Experimental results compare the performance of the algorithm...
متن کاملPerformance of Connectionist Learning Algorithms on 2-D SIMD Processor Arrays
The mapping of the back-propagation and mean field theory learning algorithms onto a generic 2-D SIMD computer is described. This architecture proves to be very adequate for these applications since efficiencies close to the optimum can be attained. Expressions to find the learning rates are given and then particularized to the DAP array procesor.
متن کامل2-d Simd Algorithms in the Perfect Shuue Networks
This paper studies a set of basic algorithms for SIMD Perfect Shuue networks. These algorithms where studied in Sc-80] and NS-81], but for the 1-D case, where the size of the problem N is the same as the number of processors P. For the 2-D case of N = L P , studied by GK-80] and Kr-81], we improve several algorithms, achieving run time O(L+log P) rather than O(L log P), as N exceeds P. We give ...
متن کامل