Meta-State Conversion

H. G. Dietz; G. Krishnamurthy

doi:10.1109/ICPP.1993.58

Meta-State Conversion

H. G. Dietz, G. Krishnamurthy

Electrical and Computer Engineering

Research output: Contribution to journal › Conference article › peer-review

6 Scopus citations

Abstract

In MIMD (Multiple Instruction stream. Multiple Data stream) execution, each processor has its awn state. Although these states are generally considered to be independent entities, it is also possible to view the set of processor states at a particular time as single, aggregate, "Meta State." Once a program has been converted into a single finite automaton based on Meta States, only a single program counter is needed. Hence, it is possible to duplicate the MIMD execution using SIMD (Single Instruction stream, Multiple Data stream) hardware without the overhead of interpretation or even of having each processing element keep a copy of the MIMD code. In this paper, we present an algorithm for Meta-State Conversion (MSC) and explore some properties of the technique.

Original language	English
Article number	4134184
Pages (from-to)	47-56
Number of pages	10
Journal	Proceedings of the International Conference on Parallel Processing
Volume	2
DOIs	https://doi.org/10.1109/ICPP.1993.58
State	Published - 1993
Event	1993 International Conference on Parallel Processing, ICPP 1993 - Syracuse, United States Duration: Aug 16 1993 → Aug 20 1993

Bibliographical note

Funding Information:
This work was supported in part by the Office of Naval Research (ONR) umder grant number N00014-91-J-4013 and by the National Science Foundation (NSF) under award number 901569-CDA.

Funding Information:
Perhaps the most obvious way to make SIMD hardware mimic MIMD execution is to write a SIMD program that will interpretively execute a MIMD instruction set. In the simplest terms, such an interpreter has a data structure, replicated in each SIMD PE, that corresponds to the internal registers of each MIMD processor. Likewise, each PE's memory holds a copy of the MIMD code to be executed. Hence, the interpreter struc¬ ture can be as simple as: * This work was supported in part by the Office of Naval Research (ONR) under grant number N0OO14-91-J-4O13 and by the National Science Foundation (NSF) under award number 9015696-CDA.

Publisher Copyright:
© 1993 IEEE.

ASJC Scopus subject areas

Software
Mathematics (all)
Hardware and Architecture

Access to Document

10.1109/ICPP.1993.58

Cite this

@article{343fb9417b3f46de8997799a92f70ce3,

title = "Meta-State Conversion",

abstract = "In MIMD (Multiple Instruction stream. Multiple Data stream) execution, each processor has its awn state. Although these states are generally considered to be independent entities, it is also possible to view the set of processor states at a particular time as single, aggregate, {"}Meta State.{"} Once a program has been converted into a single finite automaton based on Meta States, only a single program counter is needed. Hence, it is possible to duplicate the MIMD execution using SIMD (Single Instruction stream, Multiple Data stream) hardware without the overhead of interpretation or even of having each processing element keep a copy of the MIMD code. In this paper, we present an algorithm for Meta-State Conversion (MSC) and explore some properties of the technique.",

author = "Dietz, {H. G.} and G. Krishnamurthy",

note = "Funding Information: This work was supported in part by the Office of Naval Research (ONR) umder grant number N00014-91-J-4013 and by the National Science Foundation (NSF) under award number 901569-CDA. Funding Information: Perhaps the most obvious way to make SIMD hardware mimic MIMD execution is to write a SIMD program that will interpretively execute a MIMD instruction set. In the simplest terms, such an interpreter has a data structure, replicated in each SIMD PE, that corresponds to the internal registers of each MIMD processor. Likewise, each PE's memory holds a copy of the MIMD code to be executed. Hence, the interpreter struc¬ ture can be as simple as: * This work was supported in part by the Office of Naval Research (ONR) under grant number N0OO14-91-J-4O13 and by the National Science Foundation (NSF) under award number 9015696-CDA. Publisher Copyright: {\textcopyright} 1993 IEEE.; 1993 International Conference on Parallel Processing, ICPP 1993 ; Conference date: 16-08-1993 Through 20-08-1993",

year = "1993",

doi = "10.1109/ICPP.1993.58",

language = "English",

volume = "2",

pages = "47--56",

}

TY - JOUR

T1 - Meta-State Conversion

AU - Dietz, H. G.

AU - Krishnamurthy, G.

N1 - Funding Information: This work was supported in part by the Office of Naval Research (ONR) umder grant number N00014-91-J-4013 and by the National Science Foundation (NSF) under award number 901569-CDA. Funding Information: Perhaps the most obvious way to make SIMD hardware mimic MIMD execution is to write a SIMD program that will interpretively execute a MIMD instruction set. In the simplest terms, such an interpreter has a data structure, replicated in each SIMD PE, that corresponds to the internal registers of each MIMD processor. Likewise, each PE's memory holds a copy of the MIMD code to be executed. Hence, the interpreter struc¬ ture can be as simple as: * This work was supported in part by the Office of Naval Research (ONR) under grant number N0OO14-91-J-4O13 and by the National Science Foundation (NSF) under award number 9015696-CDA. Publisher Copyright: © 1993 IEEE.

PY - 1993

Y1 - 1993

N2 - In MIMD (Multiple Instruction stream. Multiple Data stream) execution, each processor has its awn state. Although these states are generally considered to be independent entities, it is also possible to view the set of processor states at a particular time as single, aggregate, "Meta State." Once a program has been converted into a single finite automaton based on Meta States, only a single program counter is needed. Hence, it is possible to duplicate the MIMD execution using SIMD (Single Instruction stream, Multiple Data stream) hardware without the overhead of interpretation or even of having each processing element keep a copy of the MIMD code. In this paper, we present an algorithm for Meta-State Conversion (MSC) and explore some properties of the technique.

AB - In MIMD (Multiple Instruction stream. Multiple Data stream) execution, each processor has its awn state. Although these states are generally considered to be independent entities, it is also possible to view the set of processor states at a particular time as single, aggregate, "Meta State." Once a program has been converted into a single finite automaton based on Meta States, only a single program counter is needed. Hence, it is possible to duplicate the MIMD execution using SIMD (Single Instruction stream, Multiple Data stream) hardware without the overhead of interpretation or even of having each processing element keep a copy of the MIMD code. In this paper, we present an algorithm for Meta-State Conversion (MSC) and explore some properties of the technique.

UR - http://www.scopus.com/inward/record.url?scp=84974560002&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84974560002&partnerID=8YFLogxK

U2 - 10.1109/ICPP.1993.58

DO - 10.1109/ICPP.1993.58

M3 - Conference article

AN - SCOPUS:84974560002

VL - 2

SP - 47

EP - 56

M1 - 4134184

T2 - 1993 International Conference on Parallel Processing, ICPP 1993

Y2 - 16 August 1993 through 20 August 1993

ER -

Meta-State Conversion

Abstract

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this