The PVM concurrent computing system: Evolution, experiences, and trends

Sunderam, Vaidy S.; Geist, G.A.; Dongarra, Jack; Manchek, Robert

doi:10.1016/0167-8191(94)90027-2

Cited by 218 publications

(72 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Our simulation program for the multithreaded version allows granularity of threads to be varied by assigning an arbitrary number of threads to each node. The scheduling of threads is software-controlled to exploit locality as much as possible rather than relying the dynamic behavior of the run-time scheduler provided by PVM or TPVM [2,7]. The measured execution time of each simulation includes the time taken for PVM processes to initialize as well as the time to execute the matrix multiplication routine.…”

Section: T M T = N T H ( P --1 ) T R + T S + T W --+ C --+ T C 8 ÷ N mentioning

confidence: 99%

“…Therefore, it is possible to build lowcost parallel machines as an alternative to more expensive Massively Parallel Processors (MPPs). For example, Parallel Virtual Machine (PVM) supports a heterogeneous parallel computing environment with message-passing [7]. In the message-passing model, a parallel machine is viewed as a collection of processors, or nodes, where the memory is disjoint and distributed among the nodes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Viability of multithreading on networks of workstations

Kwak

Liu

Hurson

1997

Lecture Notes in Computer Science

View full text Add to dashboard Cite

A b s t r a c t . Recent trend in high-performance computing focuses on networks of workstations (NOWs) as a way of realizing cost-effective parallel machines. This has been due to the availability of powerful wide-issue processors, high-speed networks, and software infrastructure systems. Due to its distributed nature, message-passing has been the choice of communication model for NOWs. This paper, however, examines the viability of using multithreading on NOWs. A matrix multiplication algorithm was studied by simulating a shared-memory abstraction on top of Parallel Virtual Machine (PVM) to characterize the behavior of multithreading. Our experiments indicate the performance of multithreading, with a small number of threads per processor, is very comparable to that of programs written using message-passing. Our studies also show multithreading has an added advantage over message-passing in that it is relatively insensitive to initial data distribution.

show abstract

Section: T M T = N T H ( P --1 ) T R + T S + T W --+ C --+ T C 8 ÷ N mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Viability of multithreading on networks of workstations

Kwak

Liu

Hurson

1997

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Parallel Virtual Machine (PVM) is a software system that supports the message passing programming paradigm [44].…”

Section: A7: Cluementioning

confidence: 99%

A Survey of the State of the Art in Performance Modeling and Prediction of Parallel and Distributed Computing Systems

Pllana¹,

Brandić²,

Benkner³

2008

IJCIR

View full text Add to dashboard Cite

Performance is one of the key features of parallel and distributed computing systems. Therefore, in the past a significant research effort was invested in the development of approaches for performance modeling and prediction of parallel and distributed computing systems. In this paper we identify the trends, contributions, and drawbacks of the state of the art approaches. We describe a wide range of the performance modeling approaches that spans from the highlevel mathematical modeling to the detailed instruction-level simulation. For each approach we describe how the program and machine are modeled and estimate the model development and evaluation effort, the efficiency, and the accuracy. Furthermore, we present an overall evaluation of the described approaches.

show abstract

“…visualize messages of a Parallel Virtual Machine [15]), used simple lines to represent both executing entities as well as the interaction among them. Irrespective of underlying software structure, lines may indicate process activity quite satisfactorily; however, a different approach must be taken when superimposing animation on top of structure.…”

Section: Snakesmentioning

confidence: 99%

Visualization of Large Nested Graphs in 3D: Navigation and Interaction

Parker¹,

Franck²,

Ware³

1998

Journal of Visual Languages & Computing

View full text Add to dashboard Cite

Most systems for visualizing large information structures use 2D graphics to view networks of nodes and arcs that represent data. To understand large structures it is often necessary to show both small-scale and large-scale structure. This has been called the problem of focus and context. Distortion, rapid zooming, elision, and multiple windows are all techniques that have been developed to provide both focus and context within single representations. We review these techniques and argue that 3D visualization has a number of advantages. A system called NestedVision3D (NV3D) will be presented that has been developed to investigate the use of 3D visualization for understanding the structure of large computer programs.NV3D is a system for visualizing large nested graphs using interactive 3D graphics. It has been tested with graphs containing more than 35,000 nodes and 100,000 relationships. We describe NV3D and its design philosophy. Basic navigation is facilitated by a set of 3D widgets, rapid scaling, and interactive elision. More experimental features include animations called snakes, which are used to trace dynamic software behavior.7/18/00 3Parker et. al.

show abstract

The PVM concurrent computing system: Evolution, experiences, and trends

Cited by 218 publications

References 4 publications

Viability of multithreading on networks of workstations

Viability of multithreading on networks of workstations

A Survey of the State of the Art in Performance Modeling and Prediction of Parallel and Distributed Computing Systems

Visualization of Large Nested Graphs in 3D: Navigation and Interaction

Contact Info

Product

Resources

About