dynamic resource allocation using virtual machines

8
DYNAMIC RESOURCE ALLOCATION USING VIRTUAL MACHINES FOR CLOUD COMPUTING ENVIRONMENT ABSTRACT Cloud computing allows business customers to scale up and down their resource usage based on needs. Many of the touted gains in the cloud model come from resource multiplexing through virtualization technology. In this paper, we present a system that uses virtualization technology to allocate data center resources dynamically based on application demands and support green computing by optimizing the number of servers in use. We introduce the concept of “skewness” to measure the unevenness in the multi-dimensional resource utilization of a server. By minimizing skewness, we can combine different types of workloads nicely and improve the overall utilization of server resources. We develop a set of heuristics that prevent overload in the system effectively while saving energy used. Trace driven simulation and experiment results demonstrate that our algorithm achieves good performance. EXISTING SYSTEM:

Upload: saibaba-gunturi

Post on 18-Aug-2015

216 views

Category:

Documents


0 download

DESCRIPTION

Abstract of Dynamic Resource Allocation Using Virtual Machines .

TRANSCRIPT

DYNAMIC RESOURCE ALLOCATION USINGVIRTUAL MACHINES FOR CLOUDCOMPUTING ENVIRONMENTABSTRACT Cloud computing allows business customers to scale up and down their resource usagebasedonneeds. Manyof thetoutedgainsinthecloudmodel comefromresourcemultiplexing through virtualization technology. In this paper, we present a system thatusesvirtualizationtechnologytoallocatedatacenterresourcesdynamicallybasedonapplication demands and support green computing by optimizing the number of serversin use. We introduce the concept of sewness! to measure the unevenness in the multi"dimensional resource utilization of a server. #y minimizing sewness, we can combinedifferent types of worloads nicely and improve the overall utilization of serverresources. We develop a set of heuristics that prevent overload in the system effectivelywhile saving energy used. $race driven simulation and experiment results demonstratethat our algorithm achieves good performance.EXISTING SYSTEM:%irtual machine monitors &%MMs' lie (en provide a mechanism for mapping virtualmachines &%Ms' to physical resources. $his mapping is largely hidden from the cloudusers. )sers with the *mazon +C, service -./, for example, do not now where their%M instances run. It is up to the cloud provider to mae sure the underlying physicalmachines &0Ms' have sufficient resources tomeet their needs. %Mlive migrationtechnologymaes it possibletochangethemappingbetween%Msand0Mswhileapplications are running.PROPOSED SYSTEM:Inthis paper, wepresent thedesignandimplementationof anautomatedresourcemanagement system that achieves a good balance between the two goal: 1verloadavoidance2 thecapacityofa0Mshouldbesufficient tosatisfytheresource needs of all %Ms running on it. 1therwise, the 0M is overloaded andcan lead to degraded performance of its %Ms. 3reen computing2 the number of 0Ms used should be minimized as long as theycan still satisfy the needs of all %Ms. Idle 0Ms can be turned off to save energy.*dvantage of 0roposed 4ystem2 We develop a resource allocation system that can avoid overload in the systemeffectively while minimizing the number of servers used. We introduce the concept of sewness! to measure the uneven utilization of aserver. #yminimizingsewness, we canimprove the overall utilizationofservers in the face of multi"dimensional resource constraints.MODULE DESCRIPTION:Number of Modue!*fter careful analysis the system has been identified to have the following modules2"# Coud Com$u%&'( Modue#)# Re!our*e M+'+(eme'% Modue#,# V&r%u+&-+%&o' Modue#.# Gree' Com$u%&'( Modue# "#Coud Com$u%&'(Modue:CloudcomputingreferstoapplicationsandservicesofferedovertheInternet. $heseservices are offered from data centers all over the world, which collectively are referredto as the 5cloud.5 Cloud computing is a movement away from applications needing to beinstalled on an individual6s computer towards the applications being hosted online. Cloudresources are usually not only shared by multiple users but as well as dynamically re"allocated as per demand. $his can wor for allocating resources to users in different timezones.)# Re!our*e M+'+(eme'% Modue:7ynamicresourcemanagement hasbecomeanactiveareaofresearchintheCloudComputing paradigm. Cost of resources varies significantly depending on configurationforusing them.8enceefficientmanagement ofresources isof primeinterest to bothCloud0roviders andCloud)sers. $hesuccess of anycloudmanagement softwarecritically de"pends on the flexibility9 scale and efficiency with which it can utilize theunderlying hardware resources while pro"viding necessary performance isolation.4uccessfulresourcemanagementsolutionforcloudenvironments, needstoprovidearich set of resource controls for better isolation, while doing initial placement and loadbalancing for efficient utilization of underlying resources.,# V&r%u+&-+%&o' Modue:%irtualization, in computing, is the creation of a virtual &rather than actual' %ersionofsomething, suchasahardwareplatform, operatingsystem, andastoragedevice or networ resources.%M live migration is a widely used techni:ue for dynamicresource allocation in a virtualized environment.$he process of running two or morelogical computersystemsoononeset ofphysical hardware.7ynamicplacement ofvirtual servers to minimize 4;* violations..# Gree'Com$u%&'( Modue:Many efforts have been made to curtail energy consumption. 8ardware basedapproaches include novel thermal design for lower cooling power, or adopting power"proportional and low"power hardware. 7ynamic %oltage and