Conceptually, the BBN Butterfly Parallel Processor can support a model of computation based on either shared memory or message passing. Recently, much of t.he work on the Butterfly, including finite element analysis and computer vision algorithms, has assumed the shared-memory model. In this paper we describe the results of our experiments with the message-passing model. The goal of the experiments was to analyze the tradeoffs between the shared-memory model, as exemplified by the BBN Uniform System package, and a simple message-passing model. Important factors to be considered were performance, scalability, and ease of programming. We compare the two models with respect to these criteria and conclude that the particular model of computatio...
This paper discusses some of the issues involved in implementing a shared-address space programming ...
Current and emerging high-performance parallel computer architectures generally implement one of two...
Clusters of Symmetrical Multiprocessors (SMPs) have recently become very popular as low cost, high p...
Conventional wisdom holds that message-passing is orders of magnitude more expensive than shared mem...
Message passing and shared memory are two techniques parallel programs use for coordination and comm...
The last decade has produced enormous improvements in processor speeds without a corresponding impro...
There has been a great deal of interest recently in the development of general-purpose bridging mode...
We compare two paradigms for parallel programming on networks of workstations: message passing and d...
This paper determines the computational strength of the shared memory abstraction (a register) emul...
. Interoperability in non-sequential applications requires communication to exchange information usi...
The benefits of hardware support for shared memory versus those for message passing are difficult to...
Shared memory is the most popular parallel programming model for multi-core processors, while messag...
This paper describes the methods used and experiences made with implementing a finite element applic...
126 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1997.It is important to study the ...
For three years, members of the Computer Science Department at the University of Rochester have used...
This paper discusses some of the issues involved in implementing a shared-address space programming ...
Current and emerging high-performance parallel computer architectures generally implement one of two...
Clusters of Symmetrical Multiprocessors (SMPs) have recently become very popular as low cost, high p...
Conventional wisdom holds that message-passing is orders of magnitude more expensive than shared mem...
Message passing and shared memory are two techniques parallel programs use for coordination and comm...
The last decade has produced enormous improvements in processor speeds without a corresponding impro...
There has been a great deal of interest recently in the development of general-purpose bridging mode...
We compare two paradigms for parallel programming on networks of workstations: message passing and d...
This paper determines the computational strength of the shared memory abstraction (a register) emul...
. Interoperability in non-sequential applications requires communication to exchange information usi...
The benefits of hardware support for shared memory versus those for message passing are difficult to...
Shared memory is the most popular parallel programming model for multi-core processors, while messag...
This paper describes the methods used and experiences made with implementing a finite element applic...
126 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1997.It is important to study the ...
For three years, members of the Computer Science Department at the University of Rochester have used...
This paper discusses some of the issues involved in implementing a shared-address space programming ...
Current and emerging high-performance parallel computer architectures generally implement one of two...
Clusters of Symmetrical Multiprocessors (SMPs) have recently become very popular as low cost, high p...