Speedup and scalability analysis of Master--Slave applications on large heterogeneous clusters

Authors:
Eduardo Javier Huerta Yero;Marco Aurélio Amaral Henriques
Affiliations:
School of Electrical and Computer Engineering, State University of Campinas, São Paulo, Brazil;School of Electrical and Computer Engineering, State University of Campinas, São Paulo, Brazil
Venue:
Journal of Parallel and Distributed Computing
Year:
2007

Citing 24
Cited 1

Reevaluating Amdahl's law

Communications of the ACM
Parallel depth first search. Part II. analysis

International Journal of Parallel Programming
Speedup Versus Efficiency in Parallel Systems

IEEE Transactions on Computers
Scalable problems and memory-bounded speedup

Journal of Parallel and Distributed Computing
Program speedup in a heterogeneous computing network

Journal of Parallel and Distributed Computing - Special issue on heterogeneous processing
Scalability of algorithms: an analytic approach

Parallel Computing
An effective and practical performance prediction model for parallel computing on nondedicated heterogeneous NOW

Journal of Parallel and Distributed Computing
Speedup and efficiency of large-size applications on heterogeneous networks

Theoretical Computer Science - Special issue on parallel computing
Performance models for scalable cluster computing

Journal of Systems Architecture: the EUROMICRO Journal - Special double issue: cluster computing
The grid: blueprint for a new computing infrastructure

The grid: blueprint for a new computing infrastructure
Performance evaluation of parallel systems

Parallel Computing - Special Anniversary issue
Evaluating the Scalability of Distributed Systems

IEEE Transactions on Parallel and Distributed Systems
A Slowdown Model for Applications Executing on Time-Shared Clusters of Workstations

IEEE Transactions on Parallel and Distributed Systems
Matrix Multiplication on Heterogeneous Platforms

IEEE Transactions on Parallel and Distributed Systems
Computer Networks

Computer Networks
SETI@home: an experiment in public-resource computing

Communications of the ACM
Scalability versus execution time in scalable systems

Journal of Parallel and Distributed Computing
Performance Metrics: Keeping the Focus on Runtime

IEEE Parallel & Distributed Technology: Systems & Technology
Performance Modeling and Prediction of Nondedicated Network Computing

IEEE Transactions on Computers
Scalability of Parallel Algorithm-Machine Combinations

IEEE Transactions on Parallel and Distributed Systems
Ten Reasons to Use Divisible Load Theory

Computer
Master/Slave Computing on the Grid

HCW '00 Proceedings of the 9th Heterogeneous Computing Workshop
The core Legion object model

HPDC '96 Proceedings of the 5th IEEE International Symposium on High Performance Distributed Computing
Foreword (Special Issue of Cluster Computing on Divisible Load Scheduling)

Cluster Computing

Scalability limits of Bag-of-Tasks applications running on hierarchical platforms

Journal of Parallel and Distributed Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Although cluster environments have an enormous potential processing power, real applications that take advantage of this power remain an elusive goal. This is due, in part, to the lack of understanding about the characteristics of the applications best suited for these environments. This paper focuses on Master/Slave applications for large heterogeneous clusters. It defines application, cluster and execution models to derive an analytic expression for the execution time. It defines speedup and derives speedup bounds based on the inherent parallelism of the application and the aggregated computing power of the cluster. The paper derives an analytical expression for efficiency and uses it to define scalability of the algorithm-cluster combination based on the isoefficiency metric. Furthermore, the paper establishes necessary and sufficient conditions for an algorithm-cluster combination to be scalable which are easy to verify and use in practice. Finally, it covers the impact of network contention as the number of processors grow.