Today's increased computing speeds allow conventional sequential machines to effectively emulate associative computing techniques. Here is a parallel programming paradigm designed for a wide range of computing engines. A ssociative computing evolved in an era when associative memories were both relatively new and, because they required a comparator at each bit of memory, relatively expensive. In the early 1970s. Goodyear Aerospace improved upon early associative processing techniques with its Staran SIMD (single instruction. multiple data) computer.' Goodyear realized that the massively parallel search capability of bit-serial SIMDs could simulate associative searching, with the cost advantage of sharing the comparison logic (that is, the processing elements) over all the bits in an entire row of memory. This approach provided two additional benefits: The word widths could be very large (from 256 bits to 64 kilobits), and the data could be processed in situ using the same PEs.However, today's lower hardware costs and increased computing speeds allow parallel techniques to be effectively emulated on conventional sequential machines. Accessing data by associative searching rather than addresses and processing data in memory require a new programming style. One goal of our research is to develop a parallel programming paradigm that is suitable for many diverse applications, is efficient to write and execute, and can be used on a wide range of computing engines, from PCs and workstations to massively parallel supercomputers.Our associative-computing (ASC) paradigm is an extension of the general associative processing techniques developed by Goodyear. We use two-dimensional tables as the basic data structure. Our paradigm has an efficient associative-based, dynamic memory-allocation mechanism that does not use pointers. It incorporates data parallelism at the base level, so that programmers do not have to specify low-level sequential tasks such as sorting, looping, and parallelization.Our paradigm supports all of the standard data-parallel and massively parallel computing algorithms. It combines numerical computation (such as convolution. matrix multiplication, and graphics) with nonnumerical computing (such as compilation, graph algorithms, rule-based systems, and language interpreters).* This article focuses on the nonnumerical aspects of ASC. The ASC modelThe ASC model is the basis of a high-level associative-programming paradigm and language. As described in the sidebar, "Properties of the ASC model," the extended model provides a basis for algorithm development and analysis similar to the

Abstract

In this paper, SIMD and MIMD solutions

Abstract

The ASC MSIMD model for parallel computation supports a generalized version of an associative style of computing that has been used since the introduction of associative SIMD computers in the early 1970's. In particular, this model supports data parallelism, constant time maximum and minimum operations, one or more instruction streams ISs which are sent to an equal number of partition sets of processors, assignment of tasks to the ISs using control parallelism. ASC also allows a network to interconnect the processing elements PEs. This paper shows how ASC can be simulated with synchronous PRAM, and the converse. These results provide an important step in de ning the power of associative model in terms of PRAM which is the most well studied parallel model. Also, these simulations will provide numerous algorithms for ASC by providing an automatic method o f c onverting algorithms from PRAM to ASC.