Static Optimization of Distributed Tuplespace Message Communications
نویسندگان
چکیده
Distributed memory parallel systems such as work station clusters stubbornly remain in need of software systems that provide programmers with an e ective but uncomplicated means of realizing improved ap plication performance Distributed shared memory systems are maturing to ll this need Tuplespace is a structured distributed shared memory that em bodies the generative communication model This pa per presents static analyses and transformations of tu plespace parallel programs that experimentally show signi cant average reductions of in the latency of using a class of tuplespace communication called mes sage tuples which seek to achieve directed communica tion within a paradigm of uncoupled communication Message tuples are identi ed statically and an alterna tive runtime communication strategy implements the improved handling of these tuples during execution
منابع مشابه
Global Compiler Analysis for Optimizing Tuplespace Communication on Distributed Systems
The generative communication paradigm of parallel programming ooers not only simplicity, but power and ex-ibility. However, the properties of associative access and uncoupled communication that give rise to this power and exibility also lie at the heart of the compiler and run-time system implementation challenges, especially on distributed memory systems. This paper provides concrete steps tow...
متن کاملTuple Counting Data Flow Analysis and its Use in Communication Optimization
Tuplespace provides parallel programmers with an abstrac tion that hides the speci c underlying architecture allowing the archi tecture to be any number of platforms ranging from shared or distributed memory to a cluster of workstations Unfortunately any abstraction of this kind necessarily introduces a trade o for the application pro grammer between ease of use and control over performance Thi...
متن کاملIssues and Experiences in Implementing a Distributed Tuplespace
Distributed memory multiprocessors and network clusters are being used increasingly as parallel computing resources due to their scalability and cost/performance advantages. However, it is generally believed that shared memory parallel programming is easier than explicit message passing programming. Although the generative communication model provides scalability like message passing and the si...
متن کاملIdentifying Tuple Usage Patterns in an Optimizing Linda Compiler
The associative tuplespace access and uncoupled communication of Linda parallel programs are the key to the power and exibility of Linda, but also lie at the heart of the compiler and run-time system implementation challenges. This paper provides concrete steps towards advanced compile-time analysis and optimization of the uncoupled communication of Linda programs. In particular, as part of an ...
متن کاملA Stable Distributed Tuple Space
An approach is presented which extends the MOMfaulttolerant implementation of the Linda model of parallel programming. The original MOM system provided persistence of tuples and tuple states across both tuplespace and worker node halt failures. Unfortunately, the requirement that system tuple space reside in a central location restricted the scalability of the MOM model. In this work, an approa...
متن کامل