27-03-2012, 05:00 PM
cluster computing full report
CLUSTER COMPUTING.docx (Size: 544.52 KB / Downloads: 51)
Abstract
The seminar report commences with an introduction to the evolution of cluster computing and hence clusters schedulers. In the beginning classification of clusters is provided. This is followed by issues involved in designing a cluster scheduler. Various time-sharing and space-sharing scheduling strategies are discussed with emphasis on Hybrid Co-scheduling, Advance Reservation, Backfilling and Pre-emptive policies. The report covers details of an open source advance cluster scheduler “Maui” as a Case Study and explains the various aspects of space-sharing policies in terms of Maui scheduler. Finally, the report concludes with a brief survey of cluster schedulers that are currently used by academia & industry and seem promising in the near future as well.
INTRODUCTION
The computing power required by applications is increasing at a tremendous rate. Hence, the search has therefore been towards devising ever faster, ever more powerful computer systems, to help tackle more and more complex problems. In addition, parallel applications have become more and more complex with increasing processing power needs induced essentially by the progress registered in many fields (telecommunication, data-mining, etc). In short we can say trend is towards high performance computing systems. Our need for computational resources in all fields of science, engineering and commerce far outstrip our ability to fulfill these needs. The usage of clusters of computers is, perhaps, one of the most promising means by which we can bridge the gap between our needs and the available resources. The usage of a COTS-based cluster system has a number of advantages including:
Why Clusters?
Since, Processor is at the heart of the computing system so this becomes the obvious choice for improvement in hardware. However, limitations due to the speed of light and due to problems of heat dissipation impose an upper bound on the capabilities of single processors. Hence, a single processor has a limit on performance and so the only way to increase performance is to do more than one computation concurrently or in parallel. Thus, today multicore CPUs are becoming more popular. However, in order to use multiple, independent processors to solve a problem in the desired time frame, a problem is to be broken up into parallel components. So called parallel programs are to be written.
Overview of Cluster
Cluster is a collection of inter-connected and loosely coupled stand-alone computers working together as a single, integrated computing resource. Clusters are commonly, but not always, connected through fast local area networks. Clusters are usually deployed to improve speed and/or reliability over that provided by a single computer, while typically being much more cost-effective than single computers of comparable speed or reliability.
Classification of Cluster High-availability (HA) clusters High-availability clusters are implemented primarily for the purpose of improving the availability of services which the cluster provides. They operate by having redundant nodes, which are then used to provide service when system components fail. HA cluster implementations attempt to manage the redundancy inherent in a cluster to eliminate single points of failure.
CLUSTER COMPUTING.docx (Size: 544.52 KB / Downloads: 51)
Abstract
The seminar report commences with an introduction to the evolution of cluster computing and hence clusters schedulers. In the beginning classification of clusters is provided. This is followed by issues involved in designing a cluster scheduler. Various time-sharing and space-sharing scheduling strategies are discussed with emphasis on Hybrid Co-scheduling, Advance Reservation, Backfilling and Pre-emptive policies. The report covers details of an open source advance cluster scheduler “Maui” as a Case Study and explains the various aspects of space-sharing policies in terms of Maui scheduler. Finally, the report concludes with a brief survey of cluster schedulers that are currently used by academia & industry and seem promising in the near future as well.
INTRODUCTION
The computing power required by applications is increasing at a tremendous rate. Hence, the search has therefore been towards devising ever faster, ever more powerful computer systems, to help tackle more and more complex problems. In addition, parallel applications have become more and more complex with increasing processing power needs induced essentially by the progress registered in many fields (telecommunication, data-mining, etc). In short we can say trend is towards high performance computing systems. Our need for computational resources in all fields of science, engineering and commerce far outstrip our ability to fulfill these needs. The usage of clusters of computers is, perhaps, one of the most promising means by which we can bridge the gap between our needs and the available resources. The usage of a COTS-based cluster system has a number of advantages including:
Why Clusters?
Since, Processor is at the heart of the computing system so this becomes the obvious choice for improvement in hardware. However, limitations due to the speed of light and due to problems of heat dissipation impose an upper bound on the capabilities of single processors. Hence, a single processor has a limit on performance and so the only way to increase performance is to do more than one computation concurrently or in parallel. Thus, today multicore CPUs are becoming more popular. However, in order to use multiple, independent processors to solve a problem in the desired time frame, a problem is to be broken up into parallel components. So called parallel programs are to be written.
Overview of Cluster
Cluster is a collection of inter-connected and loosely coupled stand-alone computers working together as a single, integrated computing resource. Clusters are commonly, but not always, connected through fast local area networks. Clusters are usually deployed to improve speed and/or reliability over that provided by a single computer, while typically being much more cost-effective than single computers of comparable speed or reliability.
Classification of Cluster High-availability (HA) clusters High-availability clusters are implemented primarily for the purpose of improving the availability of services which the cluster provides. They operate by having redundant nodes, which are then used to provide service when system components fail. HA cluster implementations attempt to manage the redundancy inherent in a cluster to eliminate single points of failure.