A parallel compute is a collection of processing elements that communicate and cooperate to solve large problems efficiently. Parallel computers vary in two fundamental architecture facets, (i) Single Instruction Multiple Data (SIMD) Vs Multiple Instruction Multiple Data (MIMD) and (ii) Shared memory Vs Distributed memory. A parallel computer with a logically shared memory system provides a single global address space to all processors, and hence a shared programming paradigm to the users. Such systems ae referred as distributed shared memory (DSM) machines.
Load balancing on DSM machines is a challenging task, even though the shared global address space may be used as a common pool for work-loads awaiting as in centralized memory systems. Accessing remote memory banks are very expensive, an appropriate distribution of work-loads across physically distributed memories helps reduce such costly remote access.
Creating parallel programs involves first decomposing the overall computation into tasks and then assigning the tasks to the processors, this step is also called as partitioning. The optimization objective for partitioning is to balance the work-load among processors and to minimize the inter process communication needs. The number of processes generated by the partitioning step may not be equal to the processors, thus a processor may be idle or loaded with multiple processes. The primary optimization objective of mapping is to balance the workload of processors and to minimize the inter-processor communication cost. Collectively, the problem of load balancing is to develop partitioning and mapping algorithm for the purpose of achieving their respective optimization objectives.
Load balancing algorithms can be broadly categorized as static or dynamic. Static load balancing algorithms distribute the processes to processors at compile time, while dynamic algorithms bind processes to processors at run time. Static load balancing algorithms rely on the estimate execution times of the processes and inter-process communication requirement. It is not satisfactory for parallel programs that are of the dynamic and/or unpredictable kind. Consequently in dynamic load balancing, processes are generated and destroyed without a pattern at run time. A dynamic load balancing algorithm consists of four components, Load Measurement rule, an Information Exchange rule, an Initiation rule and a Load Balancing Operation.
Book: Scheduling and Load Balancing in Parallel and Distributed Systems, Editors, Behrooz A. Shirazi, Krishna M. Kavi and Ali R. Hurson
Web Resource: Parallel Computing Tutorial
Power Point Presentation: Parallel Computing
If you have any questions, comments, or suggestions please do not hesitate to write a comment. We are targeting for over 1000 latest and advanced seminar topics for Computer Science, MCA and IT students.
For updates on seminar topics and other engineering related news please subscribe through email. A subscribe box is given on the top right corner.
Related posts:
hi i need to new seminar topics related to computer scince
sir please send me latest seminar topic on computer science.
i m 7th sem B.E.computer science student..plz send me recent computer science seminar topics…
sir plz give me seminar topic for mca
i need new seminar topics related to computer science.
i m last year (I.T.)engg. student please send me letest seminar
Sir,i am a final year b-tech computer science student,i’m finding out latest seminar topics,which are really challenging and not yet implemented..I’m really in need of it..
Thanking you….
sir i am M.E. student i need some useful tutorial and algorithms for parallel computing so that i can develop a good project……
sir please send me latest computerscience seminar topics sir plz plz……….sir
hai i need new seminar topic on computer science
i need seminar topics tat can be shared by four members of computer science 8th sem engineering
need seminar topics related to 8th sem computer science tat can be shared by four
i want the topic for paper presentation, and also want the topic of research work
i need a latest 2010 research topic on information technology send me as soon as possible
A parallel compute is a collection of processing elements that communicate and cooperate to solve large problems efficiently.
Parallel computers vary in two fundamental architecture facets, (i) Single Instruction Multiple Data (SIMD) Vs Multiple Instruction Multiple Data (MIMD) and (ii) Shared memory Vs Distributed memory. A parallel computer with a logically shared memory system provides a single global address space to all processors, and hence a shared programming paradigm to the users. Such systems ae referred as distributed shared memory (DSM) machines.
Load balancing on DSM machines is a challenging task, even though the shared global address space may be used as a common pool for work-loads awaiting as in centralized memory systems. Accessing remote memory banks are very expensive, an appropriate distribution of work-loads across physically distributed memories helps reduce such costly remote access.
Creating parallel programs involves first decomposing the overall computation into tasks and then assigning the tasks to the processors, this step is also called as partitioning. The optimization objective for partitioning is to balance the work-load among processors and to minimize the inter process communication needs. The number of processes generated by the partitioning step may not be equal to the processors, thus a processor may be idle or loaded with multiple processes. The primary optimization objective of mapping is to balance the workload of processors and to minimize the inter-processor communication cost. Collectively, the problem of load balancing is to develop partitioning and mapping algorithm for the purpose of achieving their respective optimization objectives.
Load balancing algorithms can be broadly categorized as static or dynamic. Static load balancing algorithms distribute the processes to processors at compile time, while dynamic algorithms bind processes to processors at run time. Static load balancing algorithms rely on the estimate execution times of the processes and inter-process communication requirement. It is not satisfactory for parallel programs that are of the dynamic and/or unpredictable kind. Consequently in dynamic load balancing, processes are generated and destroyed without a pattern at run time. A dynamic load balancing algorithm consists of four components, Load Measurement rule, an Information Exchange rule, an Initiation rule and a Load Balancing Operation.
Book: Scheduling and Load Balancing in Parallel and Distributed Systems, Editors, Behrooz A. Shirazi, Krishna M. Kavi and Ali R. Hurson
Web Resource: Parallel Computing Tutorial
Power Point Presentation: Parallel Computing
If you have any questions, comments, or suggestions please do not hesitate to write a comment. We are targeting for over 1000 latest and advanced seminar topics for Computer Science, MCA and IT students.
sir……..i need in depth on this topic,please send to my email
i m student of ms. maths with cmputer science…i want to attend de seminar on related to computer science topics….pls tel abt dat.
sir i need latest topic of computer science for seminar>>>>>>>
Hi,
please send load balancing in parallel computers seminor topic full report
hi,
i need latest seminor topics on cs plz send some topics and full report
i m n final year c.s plz send me a latest c.s related seminar topic
HI friends, i need to new technical seminor topics.
i need new seminar topics related to computer science.
i m 7th sem B.E.computer science student..plz send me recent computer science seminar topics…
sir,plz send me the latest topics regarding computerscience as soon as possible