A LITERATURE REVIEW: EFFICIENT VM MIGRATION TECHNIQUES FOR ENERGY REDUCTION IN CLOUD COMPUTING

: Broadly speaking cloud computing is nothing but a highly ‘utilitarian’ orientation of IT services where users benefited on a pay-as-you go basis. In a way it enables the hosting of pervasive applications from consumer, scientific and business domain. We expect all electronic gadgetry to be ‘energy efficient’ to possibly achievable limits. So our data centers hosting cloud application must be cost effective and the same time should avoid undue burden of carbon footprint. While excising economy on power consumption by (data center) outmost care needs to be taken so that it never at the cost services provided to end user i.e. SLA violation must be kept as low as possible. Virtualization technology is one of the key features in cloud data centers that can improve the efficiency of hardware utilization through resource sharing, migration, and consolidation of workloads. In this paper we shall cover VM Migration Algorithms for energy reduction in cloud computing along with other novel techniques.


INTRODUCTION
CLOUD computing has revolutionized the Information and Communication Technology (ICT) industry by enabling ondemand provisioning of elastic computing resources on a pay-as-you-go basis. Cloud computing is very much beneficial for small to medium organization. SME ( Small to medium enterprise) can save on up-font cost by outsourcing its computational needs to cloud provider and consequently costs of maintenance and upgrades. Second option is to buid private cloud within organization to boost effective resource managemnet and resource provisioning. Today large-scale data center contains thousands of computer nodes. These data center consumes huge amount of electric power and CO 2 (carbon dioxide) emission to environment. Given the increasing focus on reducing energy use, ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning Engineers) recently created Standard 90. 4-2016, "Energy Standard for Data Centers."(source: http://www.achrnews.com) Internet of things (IoT) is also a rapidly growing branch of IT sector [1]. IoT also makes us of cloud computing. Today large-scale data center contains thousands of computer nodes. These data center consumes huge amount of electric power and CO 2 (carbon dioxide) emission to environment. Given the increasing focus on reducing energy use, ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning Engineers) recently created Standard 90. 4-2016, "Energy Standard for Data Centers."(source: http://www.achrnews.com). Eucalyptus cloud software is widely used for creating private cloud for organization [2].

NEED OF STUDY
The only drawback with cloud computing is that it is a notorious power guzzlers and call for a stringent 'energy efficiency' regime [3]. Currently it is estimated that servers consume 0.5% of the world's total electricity usage. Server energy demand doubles every 4-6 years. With their enormous appetite for energy, today's data centers emit as much carbon dioxide as all of Argentina. Data center emissions are expected to quadruple by 2020. The average data center consumes as much energy as 25,000 households reported by Kaplan et al [4]. Between 2000 and 2007, the total power consumption of datacenters worldwide went from 70 billion to 330 billion kWh; it's projected to grow to more than 1,000 billion kWh by 2020. In 2003, the power density of a single rack of servers was between 250 W and 1.5 kW. In 2014, it had reached almost 10 kW and is projected to reach up to 30 kW by 2020. "In the actual scenario, with an average Power Usage Efficiency (PUE) of 1.8, worldwide data center energy consumption will reach 507.9 TWh by 2020", explains Mattin Grao Txapartegy, Technology & Market Analyst at Yole." Source : http://www.yole.fr

VIRTUAL MACHINE MIGRATION TECHNIQUES
The process of movement of virtual machine instances from one physical node or storage location to another is called VM migration. We are living in an age, where something may be manipulated or altered with the assistance of advance technology [5].

Live Migration and High Availability
Live migration is movement of a virtual machine instance from one physical host to anther while being powered on. Live migration is helpful in load balancing or while performing proactive maintenance in case of failure of physical machine.

Clod/Regular Migration
Cold migration can be defined as migration of powered-off or suspended virtual machine. Cold migration is simple to implement as compared to live migration. By using cold migration you can also move associated disks from one data store to another.
Before you perform VM migration to reduce energy consumption of a data center following points must be analyzed: (a)When to migrate VMs [6] Determine the best time to migrate the VM instance to reduce energy consumption without violating SLA. (b)Which VMs to migrate: once decision has been taken to perform VM migration , next step is select set VM instance for migration from one host to another; that results in effective usage of resources in cloud. (c) Where to migrate the VMs selected for migration [6]: select the set of physical machines on which VM instances to be migrated. d)When and which physical nodes to switch on/off: Last step the put the idle nodes to power saver mode or hibernation to reduce energyconsumption.

CLASSIFICATION OF VM MIGRATION ALGORITHMS FOR ENERGY REDUCTION IN CLOUD COMPUTING
Since efficient VM allocation & migration helps in energy reduction of data center, which is often modeled as bin packing problem and has been proved as NP-hard problem [7]. While excising economy on power consumption by (data center) outmost care needs to be taken so that it never at the cost services provided to end user i.e. SLA violation must be kept as low as possible.

HEURISTIC ALGORITHMS
Heuristic is a set of constraints that aim at finding a good solution for a particular problem [8]. The set of constraints used by heuristic are problem dependent and provide solution to a problem in a limitde time. These heuristic methods have various constraints like number of migrations, SLA, cost, etc. There is need to construct optimization functions in different ways. The main plus point of heuristic algorithms is that they give satisfactory solution to a problem in limited time cost frame. Heuristic algorithms are easier to implement in comparison to meta-heuristic algorithms. Since heuristic algorithms run faster, they are more suitable for online task scheduling that requires minimum response time. Greedy algorithm is a type of heuristic algorithms and is applied in the literature [9][10] [11] to quickly obtain a solution for online scheduling scenario.

META-HEURISTIC ALGORITHMS
Meta -heuristic algorithms are mainly designed for a general purpose problem. They follow uniform set of procedures to construct and solve problems. The typical meta-heuristic algorithms are bio-inspired like genetic algorithms, ACO (Ant colony optimization ,Particle Swarm Optimization and honey bee foraging algorithms.

HYBRID ALGORITHMS
In Hybrid algorithms, heuristic algorithms are used to provide initial VM placement and meta-heuristic algorithms provides optimum placement of VMs during migration. This algorithm increases the implementation complexity but reduces time and cost space. Thiruvenkadam at el [12] proposed a hybrid genetic algorithm that follows this approach.

CONCLUSION AND FUTURE DIRECTIONS
Multi tenancy, concurrency and distribution are main feature of any cloud computing architecture [52]. This paper covers different Vm Migration Algorithms for energy reduction in cloud computing (Heuristic ,Meta-heuristic& Hybrid algorithms) along with other novel techniques. The main plus point of heuristic algorithms is that they give satisfactory solution to a problem in limited time cost frame. Heuristic algorithms are easier to implement in comparison to meta-heuristic algorithms. In Hybrid algorithms, heuristic algorithms are used to provide initial VM placement and meta-heuristic algorithms provides optimum placement of VMs during migration. This algorithm increases the implementation complexity but reduces time and cost space. Depending upon the kind of problem, you can use one or other. Now, we discuss the future directions and challenges as below: • We observed that most meta-heuristic s gives better results in terms in energy consumption and QoS(Quality-of-service) but they are tested using only simulators. There is need to test these Meta-heuristic algorithms like genetic algorithms, ACO ( Ant colony optimization ,Particle Swarm Optimization and honey bee foraging algorithms in realistic environment. So that cloud vendors can adopt these methods to reduce energy consumption & ensure QoS of user's tasks. • We also have observed that how it is difficult to ensure a balance between energy consumption and SLA violations. Therefore reducing energy consumption while maintain a QoS (Quality-of-Service) is a future research challenge.
• While considering the diversity of our surveyed papers, we want to know which algorithm is best under which environment. This question still is open because of heterogeneity of different algorithms and lack of validation of these algorithms under different realistic environment. A thorough testing of all these algorithms under same environment and under heterogeneous environment is definitely required as future work.