Associative Computing Research Articles

A gentle introduction to the associative computing (ASC) model of parallel computation

Recent computing hardware technologies such as the Intel Xeon Phi and NVIDIA CUDA family of hardware coprocessors have had tremendous impact on how computer scientists program parallel computers. A...

Fine-grained parallel implementations for SWAMP+ Smith–Waterman alignment

• Extended Smith–Waterman algorithm returns multiple subalignments retaining precision. • BLAST-like results do not use heuristics, taking advantage of three architectures. • Enhanced SIMD Associative Computing Platform emulation shows linear speedup . • ClearSpeed co-processor demonstrated linear speedup on SIMD-like hardware on 96 PEs. • Convey Computer accelerator code efficiently returns non-overlapping subalignments. More sensitive than heuristic methods for searching biological databases, the Smith–Waterman algorithm is widely used but has the drawback of a high quadratic running time. The faster approach extends Smith–Waterman using Associative Massive Parallelism (SWAMP+) for three different parallel architectures: ASsociative Computing (ASC), the ClearSpeed coprocessor, and the Convey Computer FPGA coprocessor. We show that parallel versions of Smith–Waterman can be successfully modified to produce multiple BLAST-like sub-alignments while maintaining the original precision. SWAMP+ combines parallelism and the novel extension producing multiple sub-alignments for pairwise comparisons. Two parallel SWAMP+ implementations for the ASC model and the ClearSpeed CSX-620 use a wavefront approach. Both perform a full traceback in parallel memory, returning multiple sub-alignments. Results show a linear speedup for the 96 processing elements (PEs) on a single ClearSpeed chip. The third SWAMP+ adaptation uses the non-associative Convey Computer FPGA coprocessor. The hybrid system has a Smith–Waterman algorithm suite designed to produce high-speed, high-throughput alignments, optimized for large databases. The Convey Computer Smith–Waterman algorithm suite was extended to produce the additional SWAMP+ sub-alignments efficiently. The parallel sequence alignment algorithms were designed for three different computer systems, all of which contain extensions to produce multiple, additional sub-alignments. This work creates a speedup while providing a deeper exploration of the matched query sequences previously unavailable.

ASC: an associative-computing paradigm

Today's increased computing speeds allow conventional sequential machines to effectively emulate associative computing techniques. We present a parallel programming paradigm called ASC (ASsociative Computing), designed for a wide range of computing engines. 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. >

HAsP—Heterogeneous associative processing

Heterogeneous processing (HP) is a technique intended for Grand Challenge and other high performance computing (HPC) problems and for bridging the gap between the theoretical potential of parallel processing and the current reality. In HP, we aim to match code and algorithms to best-suited architectures, through techniques such as code profiling and analytic benchmarking. Associative computing (AsC) combines ideas from both associative memories and single instruction multiple data (SIMD) computers to look at new ways to use fine-grain parallel processors to achieve results beyond what is normally done by using spinoffs from sequential or multiple instruction multiple data (MIMD) processors. Heterogeneous associative processing (HAsP) is a generalization of the concepts of both HP and AsC. In HAsP, the AsC assumption of linking each datum with its own processor is generalized to assuming that each data file has its own dedicated computer. This paradigm maps onto all levels of granularity and can be easily emulated on most machines. The goal of HAsP is to allow the user to discuss the heterogeneous system at the highest possible level and with the tightest possible synchronism. HAsP offers the potential of combining the simplifying programming approaches and algorithmic efficiencies of AsC with the performance of HP.