Load Balancing Strategies for Distributed Memory Machines

Load Balancing Strategies for Distributed Memory Machines PDF Author: Ralf Diekmann
Publisher:
ISBN:
Category :
Languages : en
Pages : 37

Get Book Here

Book Description

Load Balancing Strategies for Distributed Memory Machines

Load Balancing Strategies for Distributed Memory Machines PDF Author: Ralf Diekmann
Publisher:
ISBN:
Category :
Languages : en
Pages : 37

Get Book Here

Book Description


Impact of Load Balancing on Unstructured Adaptive Grid Computations for Distributed-memory Multiprocessors

Impact of Load Balancing on Unstructured Adaptive Grid Computations for Distributed-memory Multiprocessors PDF Author: Andrew Sohn
Publisher:
ISBN:
Category : Fluid dynamics
Languages : en
Pages : 14

Get Book Here

Book Description
Abstract: "The computational requirements for an adaptive solution of unsteady problems change as the simulation progresses. This causes workload imbalance among processors on a parallel machine which, in turn, requires significant data movement at runtime. We present a new dynamic load-balancing framework, called JOVE, that balances the workload across all processors with a global view. Whenever the computational mesh is adapted, JOVE is activated to eliminate the load imbalance. JOVE has been implemented on an IBM SP2 distributed-memory machine in MPI for portability. Experimental results for two model meshes demonstrate that mesh adaption with load balancing gives more than a sixfold improvement over one without load balancing. We also show that JOVE gives a 24-fold speedup on 64 processors compared to sequential execution."

Low Contention Load Balancing on Large-scale Multiprocessors

Low Contention Load Balancing on Large-scale Multiprocessors PDF Author: Maurice P. Herlihy
Publisher:
ISBN:
Category : Multiprocessors
Languages : en
Pages : 14

Get Book Here

Book Description


Global Load Balancing with Parallel Mesh Adaption on Distributed-memory Systems

Global Load Balancing with Parallel Mesh Adaption on Distributed-memory Systems PDF Author: Rupak Biswas
Publisher:
ISBN:
Category : Cache memory
Languages : en
Pages : 24

Get Book Here

Book Description
Abstract: "Dynamic mesh adaption on unstructured grids is a powerful tool for efficiently computing unsteady problems to resolve solution features of interest. Unfortunately, this causes load imbalance among processors on a parallel machine. This paper describes the parallel implementation of a tetrahedral mesh adaption scheme and a new global load balancing method. A heuristic remapping algorithm is presented that assigns partitions to processors such that the redistribution cost is minimized. Results indicate that the parallel performance of the mesh adaption code depends on the nature of the adaption region and show a 35.5X speedup on 64 processors of an SP2 when 35% of the mesh is randomly adapted. For large-scale scientific computations, our load balancing strategy gives almost a sixfold reduction in solver execution times over non-balanced loads. Furthermore, our heuristic remapper yields processor assignments that are less than 3% off the optimal solutions but requires only 1% of the computational time."

A New Load Balancing Method For Distributed Memory Multicomputers

A New Load Balancing Method For Distributed Memory Multicomputers PDF Author: MAHEEN. ISLAM
Publisher: LAP Lambert Academic Publishing
ISBN: 9783659114977
Category :
Languages : de
Pages : 0

Get Book Here

Book Description
The load balancing algorithms are based on the intuition that for better resource utilization it is desirable for the load in distributed system to be balanced evenly. Thus, a load balancing algorithm tries to distribute the total system load among the processing elements such that each processing elements receives approximately equal amount of load. A novel distributed algorithm for load balancing is proposed here which is based on the global load balancing information and current load distribution of the system. A simulation model is developed to compare the performance of the proposed method with the previously stated methods like MCSTLB (Maximum Cost Spanning Tree Load Balancing) method, BTLB (Binary Tree Load Balancing) method, and CBTLB (Condensed Binary Tree Load Balancing) method and Cluster method. Two criteria, the execution time and the number of process migration required by different load balancing methods are used for performance evaluation. The experimental result shows that the execution time and the number of process migration required by the proposed method is better than that of the existing methods.

Load Balancing Data Parallel Programs on Distributed Memory Computers

Load Balancing Data Parallel Programs on Distributed Memory Computers PDF Author: J. de Keyser
Publisher:
ISBN:
Category : Computer programs
Languages : en
Pages : 21

Get Book Here

Book Description


Multiscale Phenomena And Their Simulation - Proceedings Of The International Conference

Multiscale Phenomena And Their Simulation - Proceedings Of The International Conference PDF Author: Frithjof Karsch
Publisher: World Scientific
ISBN: 9814546291
Category :
Languages : en
Pages : 312

Get Book Here

Book Description
Multiscale Phenomena play an essential role in the dynamics of many complex systems. Owing to their inherent nonlinearity and the involvement of many different length scales, these systems are often only be studied through numerical simulations. The book focused on common structures and problems in fluid dynamics, particle physics and macromolecule simulations. An important aspect of the discussions was the development of simulation techniques for massively parallel computers and recent advances in the construction of special purpose parallel computers.

Analysis, Architectures and Modelling of Embedded Systems

Analysis, Architectures and Modelling of Embedded Systems PDF Author: Achim Rettberg
Publisher: Springer
ISBN: 3642042848
Category : Computers
Languages : en
Pages : 326

Get Book Here

Book Description
This book presents the technical program of the International Embedded Systems Symposium (IESS) 2009. Timely topics, techniques and trends in embedded system design are covered by the chapters in this volume, including modelling, simulation, verification, test, scheduling, platforms and processors. Particular emphasis is paid to automotive systems and wireless sensor networks. Sets of actual case studies in the area of embedded system design are also included. Over recent years, embedded systems have gained an enormous amount of proce- ing power and functionality and now enter numerous application areas, due to the fact that many of the formerly external components can now be integrated into a single System-on-Chip. This tendency has resulted in a dramatic reduction in the size and cost of embedded systems. As a unique technology, the design of embedded systems is an essential element of many innovations. Embedded systems meet their performance goals, including real-time constraints, through a combination of special-purpose hardware and software components tailored to the system requirements. Both the development of new features and the reuse of existing intellectual property components are essential to keeping up with ever more demanding customer requirements. Furthermore, design complexities are steadily growing with an increasing number of components that have to cooperate properly. Embedded system designers have to cope with multiple goals and constraints simul- neously, including timing, power, reliability, dependability, maintenance, packaging and, last but not least, price.

Load Balancing: An Automated Learning Approach

Load Balancing: An Automated Learning Approach PDF Author: Pankaj Mehra
Publisher: World Scientific Publishing Company
ISBN: 981310483X
Category : Computers
Languages : en
Pages : 155

Get Book Here

Book Description
This book presents a system that learns new load indices and tunes the parameters of given migration policies. The key component is a dynamic workload generator that allows off-line measurement of task-completion times under a wide variety of precisely controlled loading conditions. The workload data collected are used for training comparator neural networks, a novel architecture for learning to compare functions of time series and for generating a load index to be used by the load balancing strategy. Finally, the load-index traces generated by the comparator networks are used in a population-based learning system for tuning the parameters of a given load-balancing policy. Together, the system constitutes an automated strategy-learning system for performance-driven improvement of existing load-balancing software.

Embedded System Design: Topics, Techniques and Trends

Embedded System Design: Topics, Techniques and Trends PDF Author: Achim Rettberg
Publisher: Springer
ISBN: 0387722580
Category : Computers
Languages : en
Pages : 456

Get Book Here

Book Description
This volume presents the technical program of the 2007 International Embedded Systems Symposium held in Irvine, California. It covers timely topics, techniques and trends in embedded system design, including design methodology, networks-on-chip, distributed and networked systems, and system verification. It places emphasis on automotive and medical applications and includes case studies and special aspects in embedded system design.