Cloud computing has revolutionized the information technology industry by enabling elastic on-demand provisioning of computing resources. Rapid growth of demand for computational power by scientific, business and web applications has led to the creation of large scale data centers consuming enormous amounts of electrical power. How to utilize this energy and how to achieve energy efficiency has been the most important issue of Green Cloud Computing.This research presents the novel technique and algorithm for the Efficient Virtual Machine Management to achieve energy efficiency. Proper use of the cloud resource by the efficient virtual machine management techniques has been carried out through the research.
Proper Virtual Machine Management is done by proper VM allocation. The overloaded host detection, VM selection ,VM placement and at last under loaded host detection are four major steps carried out throughout the research.at last shutting down the under loaded host by allocating the VM of that host to host with least power consumption is our contribution towards the research in order to achieve energy efficiency through An Efficient Virtual Machine Management.
Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction1.
The basic approach of cloud computing is computing through the Internet terminal operations that move workloads from users to the server side to share hardware, software, and information 10–16; in this way, the previous redundant wastage of resources on individual computers are avoided and the resource efficiency is greatly improved. With today’s increasingly high demand for cloud, operation of a typical cloud computing such as the large-scale data processing center evolves with a lot of power consumption. Such large amount of power consumption is contrary to today’s emphasis on energy conservation and carbon reduction, and it is a major problem that cannot be ignored. How to not only maintain the growth of the cloud computing technology, but also take into account the efficiency of energy use, is the main
research aim of this paper. The energy demand related issues cannot be ignored in the
cloud environment.in order to achieve the efficiency of energy purpose VM management can be done such that the overloaded server can be managed and at the last the VM of of underloaded host can be migrated and that can be shut down so that physical machine can be closed in order to save the energy.
In this study we propose a system for energy efficiency by VM management based on the power consumption of each host. Power consumption of each host is counted and then on base of that least power consumption host is selected as destination for migration. In that way the VM management is done.
Our solution will be helpful for the energy efficiency and VM management; through the measurement of the CPU , memory usageand power consumption. Furthermore, through energy saving algorithms of the cloud virtual machine management system, migration of virtual machines is conducted for energy saving. Finally, experimental results show that the proposed algorithm under normal usage scenarios can indeed achieve a certain degree of energy saving effect.
BACKGROUND THEORY AND RELATED SEARCH
Cloud computing is an Internet-based computing; in this way, the shared hardware and software resources and messages can be provided on demand to computers and other devices. The cloud is a metaphor for the network, or the Internet. The users do not need to know the details of the ”cloud” infrastructure or have the in-depth expertise, and are without direct control. Cloud computing allows companies to deploy applications more quickly, and reduces the complexity of management and maintenance costs to rapidly reallocate IT resources in response to business needs 10. Cloud computing describes new Internet-based services to increase IT use and easily deliver models to provide dynamic and often a virtual extension of the resource. Cloud computing can be considered as including following levels of service: Infrastructure as a Service
(IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS).Virtualization
Virtualization has emerged as vital part of Cloud Computing environment because of its capability to multiplex many Virtual Machines on the same Physical Machines, and at the same time provide isolated environment to each Virtual Machines. The software used to demultiplex the Physical Machine among many Virtual Machines is known as Virtual Machine Monitor/Hypervisor2. A Virtual Machine was originally defined by Popek and Goldberg as an efficient, isolated duplicate of a real Machine, which allows the multiplexing of the underlying Physical Machine 2. Virtualization allows fine-grained allocation of resources to Virtual Machines.
Virtualization technologies enable the execution of multiple operating system instances, or Virtual Machines on the same physical piece of hardware. Each VM works just like Physical Machines functions as if it is its own Physical Machine with a dedicated Operating System and hosted applications 2. Each VM requires access control, sometimes different between different VMs on the same Physical hardware platform. Some Virtualization platforms require an external host operating system; others are embedded directly in the hardware. There are several common approaches to Virtualization. The significant difference between the various approaches lies in the component that has visibility and control over the Virtual Machines. In some architecture, it is the hosting operating system and, in others, it is the privileged partitions 1.
To set up a standard cloud service, we need a virtual machine. When a large number of virtual machines are created through virtualization technology, it becomes very cumbersome to manage them with native instructions; hence a virtual machine management platform is needed 12,. The virtual machine management platform includes a virtual machine to create, edit, switch, pause, reply, delete, and may perform live migration operations. Next, some popular open source virtualization management platforms can be used as the network interface to provide a virtual building process with many advantages, for example, a more friendly and suitable interface is provided for monitoring states of a large number of virtual machines, and the account permissions are also easier to manage.
Rakesh kumar vishwarkarma et al. 1 represented that In order to efficiently manage the power usage of these data centers, Green computing offers schemes like load balancing across physical machines, live migration of virtual machines and Sever Consolidation which aims at minimizing the number of Active Physical Machines (APM). Server consolidation is a result of Virtual Machine (VM) scheduling which involves— VM selection, VM placement and VM placement re-optimization. He presented the VM placement optimization technique used in green cloud, particularly based on the classical problem of Bin Packing. Bin packing is inspired by the NP-Hard knapsack problem and reduces the total number of Active Physical Machines (APM). Further these placements are optimized using Rank based VM scheduling algorithm. The proposed approach subsequently reduces the energy consumption and provides improved server consolidation.
Jia-Lia Yen et al. 2 represented that As the era of information technology get into the cloud computing technology, applications are delivered as services over the Internet and the hardware and systems software in the data centers provide those services. Cloud technology might hold promise in reducing energy consumption and greenhouse effect. In the same time, the cumulative depletion of energy for the need of cloud operation is huge, especially in the large computational and data center. These servers should be controlled and used efficiently for saving energy. Therefore, he presenedt an energy management strategy of cloud. Using proposed calculation model of developed method to reduce the numbers of wake up server, they may find the way to allocate or focus on the workload of server in higher operating efficiency. And that will achieve better energy saving.
Adeleye et al. 3 represented that energy consumption of cloud’s datacentres remains the major challenge facing cloud operations and its sustainability. Efficient utilization of cloud resources using various virtual infrastructure management techniques remains one of the strategic means of realising energy efficiency in cloud. Many research works on cloud energy efficiency exist, with some focusing on the infrastructure layer and some on the virtualization layer of the cloud architecture. an attempt has been made by him to analyze various techniques deployed to manage virtual machine in an energy efficient manner. Although, the focus is on the virtualization layer, fundamental aspects such as the architectures, supporting technologies, methods, and overall cloud performances for each method will be considered.
Beloglazov et al. 4 proposed an energy efficient resource management system for virtualized Cloud data centers that reduces operational costs and provides required Quality of Service (QoS). Energy savings are achieved by continuous consolidation of VMs according to current utilization of resources, virtual network topologies established between VMs and thermal state of computing nodes. They present first results of simulation-driven evaluation of heuristics for dynamic reallocation of VMs using live migration according to current requirements for CPU performance. The results show that the proposed technique brings substantial energy savings, while ensuring reliable QoS. This justifies further investigation and development of the proposed resource management system.
Yang et al. 5 This work proposes a novel method for managing green power of a virtual machine cluster in cloud computing environments.A green power management scheme is proposed to determine how many physical machines should be run or turned off based on the gross occupies resource weight ratio of the virtual machine cluster.When the gross occupied resource weight ratio is greater than a maximum tolerant occupied resource weight ratio, a standby physical machine in the non-running physical machines is selected and waken up to join as one of the running physical machines. A resource allocation process is also used to distribute loads of the running physical machines such that the total number of the running physical
machines can be flexibly dispatched to achieve the objective of green power management.
Beloglazov et al. 6 . Dynamic consolidation of virtual machines (VMs) using live migration and switching idle nodes to the sleep mode allow Cloud providers to optimize resource usage and reduce energy consumption. However, the obligation of providing high quality of service to customers leads to the necessity in dealing with the energy-performance trade-off, as aggressive consolidation may lead to performance degradation. Due to the variability of workloads experienced by modern applications, the VM placement should be optimized continuously in an online manner. To understand the implications of the online nature of the problem, we conduct competitive analysis and prove competitive ratios of optimal online deterministic algorithms for the single VM migration and dynamic VM consolidation problems. Furthermore, we propose novel adaptive heuristics for dynamic consolidation of VMs based on an analysis of historical data from the resource usage by VMs. The proposed algorithms significantly reduce energy consumption, while ensuring a high level of adherence to the Service Level Agreements (SLA).
Accordingly, the objective of this study is to provide a technique to achieve energy efficiency .in order to achieve our aim, we used a virtual machine cluster consisting of physical machines used as host.
An efficient virtual machine management comprises the following steps:
Step 1: Calculate the total occupied resource weight of the virtual amchine cluster, which is the sum of resource weight of all the virtual machine over the sum of all available resource weight running on the host or the physical machine.The total occupied resource weight is calculated with the given equation (1):
6768941788310VMjirate=(VMjiCPUuse×VMjiRAMallocate) i=1n(VMjiCPUuse×VMjiRAMallocate) (2)
Where j is the serial number of the respective physical machine;
,i is the serial number of the virtual amchines;
N is total number of virtual amvhines;
VMjiCPUuse is processor load rate of virtual machine;
VMjiRAMallocate is memory allocation of virtual machine i in host
Step 2: After counting the resource weight of the host or physical machine select the overloaded host,the host with highest resource allocation.
4.2 PROPOSED ALGORITHM
1 For each host in hostlist doInput: Hostlist,Vmlist calculate (HOSTjrate-?):Output: Allocation of VM
VMsToMigrate.add(get VM with minimum migration time
7 For each host in hostlist do
10 For each host in hostList do
11 if host has enough resource for VM
12 if host does not get overloaded after VM migration then
14 if power;minpower then
15 allocated host?host16 minpower?power17 if allocated host?NULL then18 allocation. Add(vm,allocatedhost)
19 return allocation
20 For each host in hostlist do
21 if ishostunderloadedhost)then22 VMs to migrate. Add (host.getvmlist)
23 migrationmap. Add(getnew vm placement(VMstomigrate))
24 return migrationmap25 end if
Step 3: From the overloaded host select the VM on the basis of its migration time. The VM which takes the minimum time to migrate select that VM for migration.
Step 4: Now search for the VM placement.on which host should the VM be placed. For that count the power consumption done by each host. Now select the host with the minimum power consumption.the host on which there is the least gap of power consumption between before migration power consumption and after migration power consumption. The host with least gap select that host for VM placement. Place the selected VM on the particular host.
Now in order to fulfill our aim we have to empty the underloaded host than only the host can be shut down.
Step 5: So now find the underloaded host with minimum resource weight.
Step 6: Now select all the VM of the underloaded host.
Step 7: According to the above mentioned VM placemnt technique place the selected VM
Step 8: Now the underloaded host is empty so turn off the host in order to save the energy.
In this research four stages are important: (1) Finding overloaded host (2) VM Selection (3) VM Placement (4) Finding under loaded host.
(1) Finding overloaded host: For managing VM and getting energy efficiency in VM management first step is to find overloaded host. It is based on the idea of setting upper and lower resource utilization for hosts and keeping the total utilization of the CPU by all the VMs between these resource. If the CPU utilization of a host falls below the lower resource utilization, all VMs have to be migrated from this host and the host has to be switched to the sleep mode in order to eliminate the idle power consumption. If the utilization exceeds the upper threshold, some VMs have to be migrated from the host to reduce the utilization in order to prevent a potential SLA violation. The host with the highest resource utilization value is called to be overloaded host. Higher the value of resource utilization the value of CPU utilization will increase.
(2) VM Selection : After finding overloaded host next step is to find the host for migrating the VM. once it has been decided that a host is overloaded, the next step is to select particular VMs to migrate from this host.. After a selection of a VM to migrate, the host is checked again for being overloaded. If it is still considered as being overloaded, the VM selection policy is applied again to select another VM to migrate from the host. This is repeated until the host is considered as being not overloaded. VM selection is done considering many base like low utilization, minimum migration time, some are selected on basis of random choice ,on basis of low cpu utilization etc.
(3) VM placement : Next step after VM selection is VM placement. After selecting the VM for migration the place is to be selected that where to place the VM that is VM placement .for VM placement also many policies are taken according to the diiferent paramaeters. On what basis on an what parameter to decide the VM placement place is done in VM placement. parameters like power consumption,cost, host utilization, best fit place for VM placement. For power consumption the host where there is least power consumption is selected for VM placement.
(4) Finding Under loaded host : or determining underloaded hosts we propose a simple approach. First, all the overloaded hosts are found using the selected overloading detection algorithm, and the VMs selected for migration are allocated to the destination hosts. Then, the system finds the host with the minimum utilization compared to the other hosts, and tries to place the VMs from this host on other hosts keeping them not overloaded. If this can be accomplished, the VMs are set for migration to the determined target hosts, and the source host is switched to the sleep mode once all the migrations have been completed. If all the VMs from the source host cannot be placed on other hosts, the host is kept active. This process is iteratively repeated for all hosts that have not been considered as being overloaded.
A lot of energy can be saved by the decision in VM management which will result in reduced energy consumption, more profit to cloud service providers.
In our proposed work we add an extra parameter i.e shuting off the underloaded host, which may result into minimizing cost utilization and the numbers of running host.
In future, this proposed idea yet to be implemented and tested under real time environment
Moreover, we expect to improve energy efficiency because our goal is to have minimum power consumption.
Finally, compared to other research this will lead to significant contribution.
A lot of energy can be saved by VM management according to the algorithm proposed which will result in reduced energy consumption, more profit to cloud service providers, consequently lower price to customers and most importantly leading to Green Cloud computing environment.
The experimental setup, continuously monitor the CPU, Memory, Disk and ram allocation and calculated the power consumption in the cloud environment. The calculated power consumption helps to take strong decision and helps to reduce energy wastage.