ACM SIGMOD Record

Vineberg, Maniel

doi: 10.1145/965645.810246pmid: N/A

An algorithm to do pattern matching, a basic character string operation, is presented. The Programmable Algorithm Machine (PAM), a proposed special-purpose computer which will feature multiple processing elements and operate efficiently over a wide class of applications, is described. It is shown that the multiple processing elements of the PAM allow concurrent execution of independent operations both in a special case of the pattern matching algorithm, where the string sizes (lengths) are known at compile time, and in the general case, where the sizes are not known.

Bird, R. M.; Tu, J. C.; Worthy, R. M.

doi: 10.1145/965645.810247pmid: N/A

This paper describes an approach to solving a major problem in the information processing sciences— that of searching very large (5-50 billion characters) data bases of unstructured free-text for random queries within a reasonable time and at an affordable price. The need by information specialists and knowledge workers for large, fast low-cost text and document retrieval systems is growing rapidly. Conventional approaches to the problem have usually depended upon expensive, general purpose computers, upon special pre-preprocessing of the textual data (e.g. file inverting, indexing, abstracting, etc.), and upon elaborate, costly software. The resulting retrieval systems often cost hundreds of dollars per query and the full scanning of an uninverted, unstructured billion byte textual data base could take hours of computer services. However, in spite of these restrictions, such full text search systems have proved useful and even indispensible for many applications. Computer technology of the late 1960's and the 1970's, in both hardware and software (e.g., minicomputers, low-cost, high density disk storage, “chip” electronics, natural language query systems, etc.), have made i t practical to build special purpose, low-cost text retrieval systems. Such a system has been built, tested, and is now in a production stage. The system called the Associative File Processor (AFP), utilizes a conventional minicomputer (DEC's PDP-11/45) for control, off-the-shelf high density disks for storage, a special purpose parallel search module as a text term detector, and query and retrieval software. The AFP is currently being field tested at two sites. Full text, parallel searches on un-preprocessed textual data bases are being performed at the effective matching rates of 4 billion bytes per second (8K byte key memory times 500 Kbyte/second data stream). Estimated costs are 10 to 25 cents per query for a one billion byte data base. The costs per query and the time for searching increase in a linear fashion as data base increases. A basic architecture for the AFP is described and an implemented version is discussed. A more powerful term detector module is also under development. This system is designed around a finite state automaton algorithm.

Lipovski, G. J.

doi: 10.1145/965645.810248pmid: N/A

In analyzing two implemented intelligent secondary memories, CASSM and RAP, we recognize a common mechanism which is here called an imaginary field. The mechanism can be generalized to suggest further design possibilities. It further explains the complex and controversial CASSM mechanism, “pointer transfer” in terms of separate but related principles—imaginary fields, token transfers, and floating codes—such that further designs can utilize some or all of the techniques.

Zaky, S. G.

doi: 10.1145/965645.810249pmid: N/A

The problem of processing of non-numeric data has received considerable attention in the last few years. This is primarily motivate by the pressing needs in the are a of data base management. It has long been recognized that the parallel processing capabilities of an associative processor are fundamentally well suited to this environment. However, the complexity and cost of truely associative memories make this approach impractical. In this paper, the demands that non-numeric processing place on memory and processor hardware are discussed. Some emerging trends are presented, and a suggestion is given regarding the development of a “general purpose” microprocessor that is suited to this environment. Basically, the microprocessor discussed here is capable of performing simple search and update functions on a high speed, serial data stream. Therefore, it is suited to any application where such a situation is encountered, such as in digital communications.

Rosenthal, Robert S.

doi: 10.1145/965645.810251pmid: N/A

There has been much interest in the use of special purpose processors as the data base management component of data processing systems. The generic terms “backend” and “data management machine” have been applied to such devices. Examination of the literature reveals a broad cross section of host to backend functional distribution and interconnection methodology. This discussion represents an attempt to examine and classify several of these backend data base management machine configurations in terms of their operational parameters and application constraints. A formal taxonomy of such systems remains yet to be performed. At least three distinct classes of data management machine (DMM) are evidenced in the literature; they are the large host backend, distributed network data node and smart peripheral. The intended classes of problem that the various authors envision amenable to solution by the DMM approach exhibit overlap while the performance envelope in which each DMM architecture would provide a technically acceptable, economically sound solution to a given user requirement set varies. Some of the papers used as source for this work contained no explicit mention of either the problem classes or performance constraints that the described configuration was to address; thus liberty has been taken in interpreting the implicit application goals of these authors.

Cerretti, R.; Jasilli, D.; Matteucci, D. R.

doi: 10.1145/965645.810253pmid: N/A

The paper describes a project conducted in collaboration between CNUCE, an Institute of the Italian National Research Council, and a private research center, concerning a multifunctional system (Ulisse) capable of operating both as a stand alone unit for office administration and automation (word processsing, file management and retrieval) and as an intelligent terminal with local facilities for text editing, text formatting and storage. The first results of the application of Ulisse in the environment of the scientific users of CNUCE are illustrated.

Landson, Barry M.; Sargent, Robert G.

doi: 10.1145/965645.810255pmid: N/A

A comparison of priority queues on four different types of computer memories are made by ring a model to determine the total time to do comparable tasks. The four types of memories compared are random access (RAM), associate (AM), hybrid consisting of an associate memory and a random access memory (AM/RAM), and the hybrid memory with an auxiliary memory having the @@@@pability to perform Lewin's Associate algorithm @@@@, (AM/RAM/AML). The model used for the comparisons is an extension of the MIX model developed by Knuth 1. The MIX model was extended to include the four types of memories and the instruction set expanded include @@@@ instructions for the memories added. This @@del allows direct comparisons to be made of the different architectures and different software algorithms in performing the same tasks.