Johnnie W. Baker

ASC: an associative-computing paradigm

Todays 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.<<ETX>>

Importance of SIMD computation reconsidered

In this paper, SIMD and MIMD solutions for the real-time database management problem of air traffic control are compared. A real-time database system is highly constrained in a multiprocessor and access to the common database must be made to a limited number of data elements at a time. This MIMD database access is contrasted with the comparable SIMD common database access, which can be several hundred times greater. This is true because the SIMD can simultaneously access thousands of pertinent records instead of the limited number in the MIMD. A relatively simple example is given of a problem that has a polynomial time solution using a SIMD but for which a polynomial time solution using a MIMD is normally impossible. The fact that SIMD can support a polynomial time solution for the air traffic control problem but this problem is normally considered to be intractable for multiprocessors argues against the common belief that MIMD have greater power than SIMD. SIMD are more efficient and powerful for some critically important application areas.

Timings for associative operations on the MASC model

The MASC (Multiple Associative Computing) model is a generalized associative-style computational model that naturally supports massive data-parallelism and also control-parallelism. A wide range of applications has been developed on this model. Recent research has compared its power to the power of other popular parallel models such as the PRAM and MMB models using simulations. However, the simulation of MMB has identified some important issues regarding the cost of certain basic MASC operations required for associative computing such as broadcasts, reductions, and associative searches. This paper investigates these issues and gives background information and an analysis of timings for these operations, based on implementation techniques and comparison fairness with respect to other models. It aims to provide justification and clarify arguments on the timings for these constant-time or nearly constant-time basic MASC operations.

An associative static and dynamic convex hull algorithm

This paper presents a new static and dynamic recursive parallel algorithm for the convex hull problem. This algorithm is a parallel adaptation of the Graham Scan and Quick Hull algorithms. The computational model selected for this algorithm is the associative computing model (ASC) which supports massive parallelism through the use of data parallelism and constant time associative search and maximum functions. Also, ASC can be supported on existing SIMD computers. The static algorithm requires O(n) space, O(log n) average case running time, and O(n) worst case running time. If O(log n) ISs are used the, static algorithm should have an average running time of O(log log n).

A multiple associative model to support branches in data parallel applications using the manager-worker paradigm

ASC (associative computing model) and MASC (multiple associative computing model) have long been studied in the Department of Computer Science at Kent State University. While the previous studies provide the background and the basic definition of the model, the description of the interactions between the instruction streams (ISs) is very brief, high level, and incomplete. One change here is that we specify the interaction between ISs and consider that all of the ISs operate on the same clock in order to support predictable worst case computation times, while earlier the ISs were assumed to interact in a MIMD type fashion. This paper provides a detailed explanation as to how these interactions can be supported in the case where only a few ISs are supported.

Multiple instruction stream control for an associative model of parallel computation

This paper describes a system software design for multiple instruction stream control in a massively parallel associative computing environment. The purpose of providing multiple instruction stream control is to increase throughput and reduce the amount of parallel slackness inherent in single instruction stream parallel programming constructs. The multiple associative computing (MASC) model is used to describe this technique and a brief introduction to the MASC model of parallel computation is presented. A simple parallel computing example is used to illustrate the techniques for multiple instruction stream control in a massively parallel runtime environment.

Simulating PRAM with a MSIMD model (ASC)

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 1970s. In particular, this model supports data parallelism, constant time maximum and minimum operations, one or more instruction streams (ISs) which are sent to a unique set in a dynamic partition of the processors, and 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 simulate synchronous PRAM, and the converse. These results provide an important step in defining the power of associative model in terms of PRAM which is the most studied parallel model. Also, these simulations will provide numerous algorithms for ASC by giving an automatic method of converting algorithms from PRAM to ASC.

SWAMP: Smith-Waterman using associative massive parallelism

One of the most commonly used tools by computational biologists is some form of sequence alignment. Heuristic alignment algorithms developed for speed and their multiple results such as BLAST [1] and FASTA [2] are not a total replacement for the more rigorous but slower algorithms like Smith- Waterman [3]. The different techniques complement one another. A heuristic can filter dissimilar sequences from a large database such as GenBank [4] and the Smith-Waterman algorithm performs more detailed, in-depth alignment in a way not adequately handled by heuristic methods. An associative parallel Smith-Waterman algorithm has been improved and further parallelized. Analysis between different algorithms, different types of file input, and different input sizes have been performed and are reported here. The newly developed associative algorithm reduces the running time for rigorous pairwise local sequence alignment.

Relating the power of the Multiple Associative Computing (MASC) model to that of reconfigurable bus-based models

The MASC (Multiple ASsociative Computing) model is a multi-SIMD model that uses control parallelism to coordinate the interaction of data parallel threads and supports associative SIMD computing on each of its threads. There have been a wide range of algorithms developed for this model. Research on using this model in real-time system applications and building a scalable MASC architecture is currently quite active. In this paper, we present simulations between MASC and reconfigurable bus-based models, e.g., various versions of the Reconfigurable Multiple Bus Machine (RMBM). Constant time simulations of the basic RMBM by MASC and vice versa are obtained. Simulations of the segmenting RMBM, fusing RMBM, and extended RMBM by MASC in non-constant time are also discussed. By taking advantage of previously established relationships between RMBM and two other popular parallel computational models, namely, the Reconfigurable Mesh (RM) and the Parallel Random Access Machine (PRAM), we extend our simulation results to further categorize the power of the MASC model in relation to RM and PRAM.

Comparisons of air traffic control implementations on an associative processor with a MIMD and consequences for parallel computing

This paper has two complementary focuses. The first is the system design and algorithmic development for air traffic control (ATC) using an associative SIMD processor (AP). The second is the comparison of this implementation with a multiprocessor implementation and the implications of these comparisons. This paper demonstrates how one application, ATC, can more easily, more simply, and more efficiently be implemented on an AP than is generally possible on other types of traditional hardware. The AP implementation of ATC will take advantage of its deterministic hardware to use static scheduling. The software will be dramatically smaller and cheaper to create and maintain. Likewise, a large AP system will be considerably simpler and cheaper than the MIMD hardware currently used. While APs were used for ATC-type applications earlier, these are no longer available. We use a ClearSpeed CSX600 accelerator to emulate the AP solutions of ATC on an ATC prototype consisting of eight data-intensive ATC real-time tasks. Its performance is compared with an 8-core multiprocessor (MP) using OpenMP. Our extensive experiments show that the AP implementation meets all deadlines while the MP will regularly miss a large number of deadlines. The AP code will be similar in size to sequential code for the same tasks and will avoid all of the additional support software needed with an MP to handle dynamic scheduling, load balancing, shared resource management, race conditions, false sharing, etc. At this point, essentially only MIMD systems are built. Many of the advantages of using an AP to solve an ATC problem would carry over to other applications. AP solutions for a wide variety of applications will be cited in this paper. Applications that involve a high degree of data parallelism such as database management, text processing, image processing, graph processing, bioinformatics, weather modeling, managing UAS (Unmanned Aircraft Systems or drones) etc., are good candidates for AP solutions. This raises the issue of whether we should routinely consider using non-multiprocessor hardware like the AP for applications where substantially simpler software solutions will normally exist. It also raises the question of whether the use of both AP and MIMD hardware in a single hetergeneous system could provide more versatility and efficiency. Either the AP or MIMD could serve as the primary system, but could hand off jobs it could not handle efficiently to the other system.