IJRIT International Journal of Research in Information Technology, Volume 3, Issue 3, March 2015, Pg. 17-21

International Journal of Research in Information Technology (IJRIT)

www.ijrit.com

ISSN 2001-5569

Mixed Priority Elastic Resource Allocation in Cloud Computing Environment K. Santhi1, M.Deepa2, C.Priyadarshini3 1,2

Assistant Professor(Sr), School of Information Technology, VIT University, Tamilnadu, India.

3

School of Information Technology, VIT University, Tamilnadu, India. 1

[email protected], [email protected], [email protected] ABSTRACT

Cloud computing is a distributed computing over a network, and means the ability to run a program or application on many connected computers at the same time. The phrase also more commonly refers to networkbased services, which appear to be provided by real server hardware, and are in fact served up by virtual hardware, simulated by software running on one or more real machines. Such virtual servers do not physically exist and can therefore be moved around and scaled up (or down) on the fly without affecting the end user arguably, rather like a cloud. In private cloud during registration large amount of client are trying to register at same time from different places so heavy load is going to server due to this heavy load to resources the server unable to give response to client which lead to the problem client can’t register properly within given time or wait for long time until server recover from this problem. This problem can be solved by allocating the resources properly to server this comes under the infrastructure as a service (IaaS) where we can use and allocate a various kind of resources. To solve this problem efficiently according to virtual machine are allocated based on mixed priority scheduling where client can fix their own priority for their task so that the client need not to wait for long time and they can register properly within a short time. We have implemented in CloudSim simulator and show the result.

Keywords: - Mixed priority scheduling, IaaS, CloudSim

1. Introduction In private cloud during registration large amount of client are trying to register at same time from different places so heavy load is going to server due to this heavy load to resources the server unable to give response to client which lead to the problem client can’t register properly within given time or wait for long time until server recover from this problem.

K. Santhi, IJRIT- 17

IJRIT International Journal of Research in Information Technology, Volume 3, Issue 3, March 2015, Pg. 17-21

The main motivation in this paper the client can’t register properly within given time or wait for long time until server recover from heavy load problem. Hence if client can’t register means they can’t work in that site future and company may loss many client because of heavy load problem.

2. Related works: Rajnikant B. Wagh et. el [1] use the modified waiting queue for allocating the resource based on SLA given by user. In this they use the stack to store user request and pop the stack when they need. They use the dynamic priority scheduling algorithm and cloud min-min scheduling where the server set the priority for task based on load in server so in this user can’t set their own priority for their task. Yuuki et. el.[2] allocate the resources based on joint multiple resource allocation method which reduce the different network delay to users at multiple locations. They calculate the network delay as long delay and short delay and then they allocate the resource hence it properly reduce request loss. But they didn’t concentrate on scheduling the resource based on user request and also SLA. V.Goswami et. el[3] they use the finite multiserver queuing model with queue dependent heterogeneous servers where the web applications are modeled as queues and the virtual machines are modeled as service providers. Cloud Computing Service Provider can use multiple servers and the number of busy servers changes depending on the queue length for reducing queue length and waiting time. This helps us to dynamically create and remove virtual machines in order to scaling up and down. Zhiyuan Wang et. el. [4] use the priority scheduling algorithm for allocating virtual machine in cloud environment and they divide the work based on size of work if the user having large work means that user will get more priority then other. Then that user will be allocate first to the host that time other user who having small or medium work need to wait for long time. Jan-Jan wu et. el. [5] use the dynamic round robin algorithm to reduce the power consumption problem in datacenter by reducing the load in server by allocating the virtual machine to the server they measuring the peak power based on a particular ideal power it is static method to measure and also they does mention about turn round time, waiting time, response time.

3. Implementation 1. Get SLA data. 2. Implementing MUF Algorithm. 3. Creating VM. 4. Implementing priority scheduling algorithm. 5. Executing CloudSim. 6. Service accepts by user. 3.1. Get SLA data As soon as client connect with server the server will give user id for each client then it get the basic Service Level Agreement (SLA) details from each user that are size they required to store the data. Criticality

K. Santhi, IJRIT- 18

IJRIT International Journal of Research in Information Technology, Volume 3, Issue 3, March 2015, Pg. 17-21

for their task to complete in server. Deadline time to calculate the laxity for the user. Laxity is CPU time and current time is subtracted from deadline time. User priority from the user for their task in server. 3.2. Implementing MUF Algorithm The Maximum Urgency First scheduling algorithm is a kind of mixed priority scheduling; it uses both fixed and dynamic scheduling. The urgency priority is based on two fixed priorities and one dynamic priority. Benefit of MUF: •

Context switch takes 66 µs and a rescheduling operation in the dynamic scheduler takes 26 µs on a MC680x0 processor



It is also possible to set the priorities so that one will have control over which tasks will fail in the first place if the CPU gets overloaded.



The MUF scheduling algorithm will make sure that no critical tasks will fail their deadlines.

Criticality The first fixed priority is the criticality. This is set to either high or low and it tells if this particular task is critical to finish within its deadline. User Priority The other fixed priority is the user priority. This gives a chance for the user to set which tasks are more important not to fail. Minimum Laxity Priority The dynamic priority is a reverse version of laxity. laxity = deadline_time – current_time – CPU_time_still_needed Actions by the MUF Scheduler in Run-Time 1. Choose the task with highest criticality first. If there is more than one to choose here, go to tep 2. 2. Choose the task with the lowest laxity. If there is more than one to choose here, go to step 3 3. Choose the task with the highest user priority. If there is more than one to choose here, go to step 4. 4. Use “first-come-first-serve” to choose which task to run. 3.3. Creating VM VM having the parameters as id, user id, MIPS, number of pes(amount of CPU), amount of RAM, amount of bandwidth, amount of storage, virtual machine monitor, cloudlet scheduler. We need to create the cloudlet for VM the cloudlet is an extension to the cloudlet. It stores, despite all the information encapsulated in the Cloudlet, the ID of the VM running it. The broker needs to submit the VM and cloudlet to cloudsim. The

K. Santhi, IJRIT- 19

IJRIT International Journal of Research in Information Technology, Volume 3, Issue 3, March 2015, Pg. 17-21

DatacentreBroker represents a broker acting on behalf of a user. It hides VM management, as vm creation, submission of cloudlets to this VMs and destruction of VMs.

4. Implementing priority scheduling algorithm While executing VM in cloudsim that time the VM automatically scheduled based on maximum MIPS as a first priority then other VM. CloudletSchedulerTimeShared implements a policy of scheduling performed by a virtual machine. Cloudlets execute time-shared in VM. VmSchedulerTimeShared is a VMM allocation policy that allocates one or more Pe to a VM, and allows sharing of PEs by multiple VMs.

5. Executing CloudSim To execute the cloudsim we need to create the Datacenter class is a CloudResource whose hostList are virtualized. It deals with processing of VM queries (i.e., handling of VMs) instead of processing Cloudletrelated queries. So, even though an AllocPolicy will be instantiated (in the init() method of the superclass, it will not be used, as processing of cloudlets are handled by the CloudletScheduler and processing of VirtualMachines are handled by the VmAllocationPolicy. Host executes actions related to management of virtual machines (e.g., creation and destruction). A host has a defined policy for provisioning memory and bw, as well as an allocation policy for Pe's to virtual machines. A host is associated to a datacenter. It can host virtual machines.

6. Service accepts by user Finally the user will get service as they requested in SLA and they will be registered soon with help of virtual machine. The virtual machine is created based on the user id given to user at first.

7. Screen shots Client giving SLA details

Fig 1.Client giving SLA details

Fig 2 Server executing CloudSim

K. Santhi, IJRIT- 20

IJRIT International Journal of Research in Information Technology, Volume 3, Issue 3, March 2015, Pg. 17-21

Fig 3 Client registers successfully

4. Conclusion Using MUF and Priority Scheduling algorithm VM allocation is efficient and elastic in cloud environment. It reduce heavy load to server and make easy to register to user. It take very less time than Round Robin Scheduling algorithm since it take minimum context switch takes 66 µs and a rescheduling operation in the dynamic scheduler takes 26 µs on a MC680x0 processor. It provide the high QoS(Quality of Service) since based on client giving SLA server allocating VM.

5. References [1]. Chandrashekhar S. Pawar, Rajnikant B. Wagh “Priority Based Dynamic Resource Allocation in Cloud Computing with Modified Waiting Queue” IEEE International Conference on Intelligent Systems and Signal Processing (ISSP) on 2013. [2]. Yuuki AWANO and Shin-ichi KURIBAYASHI “A joint multiple resource allocation method for cloud computing environments with different QoS to users at multiple locations” IEEE International Conference on Utility and Cloud Computing on 2013. [3]. V. Goswami, S. S. Patra, G. B. Mund “Performance Analysis of Cloud with QueueDependent Virtual Machines” IEEE International Conference on Recent Advances in Information Technology on 2013. [4]. Jing Xiao, Zhiyuan Wang “A Priority based Scheduling Strategy for Virtual Machine Allocation in Cloud Computing Environment” International Conference on Cloud Computing and Service Computing on 2012. [5]. Jan-Jan Wu, Pangfeng Liu, Ching-Chi Lin “Energy-efficient Virtual Machine Provision Algorithms for Cloud Systems” Fourth IEEE International Conference on Utility and Cloud Computing on 2012.

K. Santhi, IJRIT- 21

Mixed Priority Elastic Resource Allocation in Cloud Computing ... - IJRIT

Cloud computing is a distributed computing over a network, and means the ... In this they use the stack to store user request and pop the stack when they need.

815KB Sizes 8 Downloads 84 Views

Recommend Documents

Implementation of Cloud Computing in remote Learning - IJRIT
Key words: Cloud computing, IaaS, SaaS, PaaS. 1. INTRODUCTION. Post-freedom time has seen India thrive surprisingly in the field of giving higher training.

Security and Interoperability in Cloud Computing and Their ... - IJRIT
online software applications, data storage and processing power. ... Interoperability is defined as Broadly speaking, interoperability can be defined ... Therefore, one of the solutions is to request required resources from a cloud IaaS provider.

PRedictive Elastic ReSource Scaling for cloud ... - Research at Google
(1) deciding how much resource to allocate is non-trivial ... 6th IEEE/IFIP International Conference on Network and Service Management (CNSM 2010),.

FAIRNESS OF RESOURCE ALLOCATION IN CELLULAR NETWORKS
1. Introduction. Fairness is an important property of a resource allocation .... for wireless packet services (see e.g., [11, 12, 13]), the Gilbert-Elliot model [14, 15].

Elastic computing with R and Redis - GitHub
May 16, 2016 - Listing 3 presents a parallel-capable variation of the boot function from the ... thisCall