A Data Flow Computer Architecture with Program and Token Memories

Sowa,; Murata,

doi:10.1109/tc.1982.1676097

Cited by 13 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In view of data flow parallelism, Zehendner et al [24] proposed three levels to be exploited: task, block and instruction, and subinstruction. Despite tireless efforts by researchers to innovate better and faster performance data flow computers [2,3,4,5], major issues on sensitive operations such as language and data dependencies, token matching, resource management, and well-formed parallelism [6] still persisted.…”

Section: Function2mentioning

confidence: 99%

The N-Dimension Computing Machine Postulate

Sophatsathit

2020

IEEE Access

View full text Add to dashboard Cite

This paper postulates a novel N-dimension computing machine that operates in an unconventional manner. This postulate aims at solving existing problems in higher dimensions , where one must rethink the scope of a given problem domain beyond the one-dimension Turing machine to dictate all subsequent problem representation, problem transformation, and algorithmic derivation. Two oversimplified well-known problems, namely, the Traveling Salesman Problem and the Tower of Hanoi problem are presented to demonstrate the point. Both synthetic problems are effectively adapted to solve a real world project. To realize the postulate in a viable architectural construct, data flow and molecular computers are investigated since they show potential computation power. Unfortunately, they are still confined to working in one-dimension domain. A biological-like architecture for software systems is proposed in three design aspects: structure, function, and behavior. Contributions of this work are to revamp traditional Turing computation paradigm to N-dimension computing machine, yet it is simple, straightforward, and implementable by state-of-the-practice hardware and software technologies. Thus, the burden of solving difficult problems can be lessened.

show abstract

Section: Function2mentioning

confidence: 99%

The N-Dimension Computing Machine Postulate

Sophatsathit

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This network partitioning, however, requires loading graphs in a way that balances workload among PE's to Avoid instruction-execution con-[icts. Sows and MurAtA have proposed partitioning the common memory into program memory and token memory to reduce the size and complexity of the switching networks [20]. Another Approach is to eliminate the arbitration network entirely by Attachins A dedicated PE to each CB.…”

mentioning

confidence: 99%

Graph allocation in static dataflow systems

Hong

Payne

Ferguson

1986

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

One of the most Important considerations for a dataflow multlprocessor is the algorithm by which the nodes of a program Ipmph are allocated for execution to its processors. In the case of the static type of architecture one must consider pipelining as well as spatial concurrency. This paper uses a graph partition.lag scheme to aid in the design of such algorithms and selection of the associated interconnection topology. The scheme first reflees the usual dataflow program graph and then partitions it into "trees? Our study shows that the hypercube interconnection a¢-¢ommodates these trees in a way that allows us to develop an algozlthm that places producer nodes (of a dataflow graph) nears their ¢onsume~s to keep the message path very short. The algorithm at~leves concurrency in simultaneous execution of subgraphs that involve parallel operations, e.g., array operations, and subgraphs ~t are highly Iteratlve or recurslve. An analytical model is given to evaluate the performance of the algorlthm.

show abstract

“…It does so by accessing [SM82] template "DOT" (Figure 2). "Apply" then equates the template's local name [Y with the value of its second input operand, the value "A", and similarly equates w to be 'Z."…”

Section: The Execution Modelmentioning

confidence: 99%

“…In this paper we will distinguish functional languages, which are generally recursive in proyramming style (being rooted in the lambda calculus [Bar841 [Dar85]), from dataflow languages [Ack82] which, having their underlying foundations in the Petri net [Rei851 have cyclic program graphs [Den751 [GKWSS] and therefore an iterative programming style [Ack82]) as wcll as other sequential biases, such as a preoccupation with time-sequential "streams" [CDP83] IKro831 Wit891. The dataflow approach has dominated recent research into datadriven parallel machine architectures [GKWSS] [SM82] [SY+89] fTp+91], while advanced architectures for functional languages have generally been based on demanddriven ("lazy," that is, normal-order evaluation) implementations [Cur773. But the data-driven functional programming style has the potential of exposing a tremendous amount of inherent parallelism in a broad class of vector and matrix oriented computations [Veg84], something that has only been seriously investigated for the dataflow paradigm [AA821 [CA88].…”

Section: Introductionmentioning

confidence: 99%

A functional execution model for a non-dataflow tagged token architecture

Jennings

Proceedings Sixth International Parallel Processing Symposium

View full text Add to dashboard Cite

We propose a new execution model for a non-dataflow tagged-token architecture which is not Petri-net based but rather more closely related to the lambda calculus. The model exploits a functional programming style having applicative-order evaluation. The computation's execution graph is dynamically generated according to easily understood dynamic tagging rules which have been demonstrated to be implementable. The model permits conceptually unboundedparallelism for an interesting class of list-oriented computations. We explain the model with the help of a simple dot-product computation as an example. We highlight some of the major differences between the dataflow paradigm and our own. Architectural issues toward implementation are briefly discussed.

show abstract

A Data Flow Computer Architecture with Program and Token Memories

Cited by 13 publications

References 6 publications

The N-Dimension Computing Machine Postulate

The N-Dimension Computing Machine Postulate

Graph allocation in static dataflow systems

A functional execution model for a non-dataflow tagged token architecture

Contact Info

Product

Resources

About