Virtual Machine Migration in Cloud Computing Environments

Modern Cloud Data Centers exploit virtualization for efficient resource management to reduce cloud computational cost and energy budget. Virtualization empowered by virtual machine (VM) migration meets the ever increasing demands of dynamic workload by relocating VMs within Cloud Data Centers. VM migration helps successfully achieve various resource management objectives such as load balancing, power management, fault tolerance, and system maintenance. However, being resourceintensive, the VM migration process rigorously affects application performance unless attended by smart optimization methods. Furthermore, a Cloud Data Centre exploits server consolidation and DVFS methods to optimize energy consumption. This paper reviews state-of-the-art bandwidth optimization schemes, server consolidation frameworks, DVFS-enabled power optimization, and storage optimization methods over WAN links. Through a meticulous literature review of state-of-the-art live VM migration schemes, thematic taxonomies are proposed to categorize the reported literature. The critical aspects of virtual machine migration schemes are investigated through a comprehensive analysis of the existing schemes. The commonalties and differences among existing VM migration schemes are highlighted through a set of parameters derived from the literature. Finally, open research issues and trends in the VM migration domain that necessitate further consideration to develop optimal VM migration schemes are highlighted.

ACM SIGAPP Applied Computing Review, 2018

Maximizing the number of virtual infrastructures spawned out of a data center is a prime concern of cloud service providers to improve their revenue and the customers' quality-of-experience. Optimal placement of topology sensitive virtual data centers on its physical counterparts increases the resource utilization and thereby the revenue. However, provisioning, scaling, and de-provisioning of virtual data centers, over the time, leaves the cloud data center fragmented. Embedding new virtual data centers on a fragmented data center would require migration of virtual machines and virtual network, already mapped to it. However, these migrations are costly in terms of the resource usage and possible violations of the terms of service level agreement. The costs of these migrations associated with a new embedding should be controlled to increase the profit. In this paper, the problem of finding virtual machine migrations with minimum cost is formally defined and two meta-heuristic sol...

2016

Virtualization technology reduces cloud operational cost by increasing cloud resource utilization level. The incorporation of virtualization within cloud data centers can severely degrade cloud performance if not properly managed. Virtual machine (VM) migration is a method that assists cloud service providers to efficiently manage cloud resources while eliminating the need of human supervision. VM migration methodology migrates current-hosted workload from one server to another by either employing live or non-live migration pattern. In comparison to non-live migration, live migration does not suspend application services prior to VM migration process. VM migration enables cloud operators to achieve various resource management goals, such as, green computing, load balancing, fault management, and real time server maintenance. In this paper, we have thoroughly surveyed VM migration methods and applications. We have briefly discussed VM migration applications. Some open research issues...

Journal of Network and Computer Applications, 2015

Modern Cloud Data Centers exploit virtualization for efficient resource management to reduce cloud computational cost and energy budget. Virtualization empowered by virtual machine (VM) migration meets the ever increasing demands of dynamic workload by relocating VMs within Cloud Data Centers. VM migration helps successfully achieve various resource management objectives such as load balancing, power management, fault tolerance, and system maintenance. However, being resourceintensive, the VM migration process rigorously affects application performance unless attended by smart optimization methods. Furthermore, a Cloud Data Centre exploits server consolidation and DVFS methods to optimize energy consumption. This paper reviews state-of-the-art bandwidth optimization schemes, server consolidation frameworks, DVFS-enabled power optimization, and storage optimization methods over WAN links. Through a meticulous literature review of state-of-the-art live VM migration schemes, thematic taxonomies are proposed to categorize the reported literature. The critical aspects of virtual machine migration schemes are investigated through a comprehensive analysis of the existing schemes. The commonalties and differences among existing VM migration schemes are highlighted through a set of parameters derived from the literature. Finally, open research issues and trends in the VM migration domain that necessitate further consideration to develop optimal VM migration schemes are highlighted.

Resource Management and Efficiency in Cloud Computing Environments

Cloud computing service has been on the rise over the past few decades, which has led to an increase in the number of data centers, thus consuming more amount of energy for their operation. Moreover, the energy consumption in the cloud is proportional to the resource utilization. Thus consolidation schemes for the cloud model need to be devised to minimize energy by decreasing the operating costs. The consolidation problem is NP-complete, which requires heuristic techniques to get a sub-optimal solution. The authors have proposed a new consolidation scheme for the virtual machines (VMs) by improving the host overload detection phase. The resulting scheme is effective in reducing the energy and the level of Service Level Agreement (SLA) violations both, to a considerable extent. For testing the performance of implementation, a simulation environment is needed that can provide an environment of the actual cloud computing components. The authors have used CloudSim 3.0.3 simulation toolkit that allows testing and analyzing Allocation and Selection algorithms.

2007

Virtual Machines are becoming increasingly valuable to resource consolidation and management, providing efficient and secure resource containers, along with desired application execution environments. This paper focuses on the VM-based resource reservation problem, that is, the reservations of CPU, memory and network resources for individual VM instances, as well as for VM clusters. In particular, it considers the scenario where one or several physical servers need to be vacated to start a cluster of VMs for dedicated execution of parallel jobs. VMs provide a primitive for transparently vacating workloads through migration; however, the process of migrating several VMs can be timeconsuming and needs to be estimated. To achieve this goal, this paper seeks to provide a model that can characterize the VM migration process and predict its performance, based on a comprehensive experimental analysis. The results show that, given a certain VM's migration time, it is feasible to predict the time for a VM with other configurations, as well as the time for migrating a number of VMs. The paper also shows that migration of VMs in parallel results in shorter aggregate migration times, but with higher per-VM migration latencies. Experimental results also quantify the benefits of buffering the state of migrated VMs in main memory without committing to hard disks.

Journal of Parallel and Distributed Computing, 2017

Cloud computing is a model for providing computing resources as a utility which faces several challenges on management of virtualized resources. Accordingly, virtual machine placement and migration are crucial to achieve multiple and conflicting goals. Regarding the complexity of these tasks and plethora of existing proposals, this work surveys the state-of-the-art in the area. It presents a cloud computing background, a review of several proposals, a discussion of problem formulations, advantages and shortcomings of reviewed works. Furthermore, it highlights the challenges for new solutions and provides several open issues, showing the relevancy of the topic in an increasing and demanding market.

International Journal of Emerging Trends in Engineering Research, 2020

Generally, the architecture of a cloud computing system contains several layers, which includes hardware layer, infrastructure layer, Platform layer, and the Services layer. Each layer performs different kinds of operations and provides different types of services. VM Migration is undertaken by a scheduler installed within computing nodes with the main aim of balancing the Load on Physical servers. The Migration is undertaken based on the internal design of a cloud computing system. The response time agreed with the user and contained within an SLA gets compromised due to the heavy processing required for deciding on the migration and actually effecting the Migration. Unlike a traditional Virtual Machine (VM), a container is an emerging lightweight virtualization technology that operates at the operating system level to encapsulate a task and its library dependencies for execution. In this paper, a modified layered architecture is presented which considers the addition of another layer (Optimised Migration Layer) that is responsible for optimising the process of migration focusing on reduction of delay time caused due effecting the migration. The additional layer uses the concept of containerisation for optimising and reduction in delay time caused due to effecting the Migration. An efficient placement algorithm has been implemented in the OML layer. OML includes certain components, such as validation, business, transformation, and the deployment phase. The component that is responsible for effecting the Migration is quarantined by making changes to the configuration files. OML is designed using four phases that include Validation, Business phase, Transformation phase. The scheduling algorithm calls the software situated in OML layer which checks the validity of the request and the actual migration is effected in the Business phase in which an efficient Migration/Placement algorithm is implemented. Here, the placement algorithm, named Squirrel Whale Optimization Algorithm (S-WOA) is implemented. The migration is done based on five conditions, which include VM to Physical Machine (PM), container to VM under same PM, and the container to VM under different PM, tasks in VM to container, and container tasks to VM. A fitness function is designed and implemented to find, best migration method. The fitness function is designed using several parameters that include bandwidth, resource utilization and load. In the transformation phase the transformation from VM to container and vice versa is undertaken. The performance of the migration strategy in cloud based on S-WOA is evaluated in terms of number of instantiated VMs, CPU utilization, memory utilization, number of activated PMs, and time. The proposed S-WOA method achieves the maximal CPU utilization of 0.483, maximal memory of 0.523, and minimal time of 4.99sec.

The emerging cloud computing paradigm provides administrators and IT organizations with tremendous freedom to dynamically migrate virtualized computing services between physical servers in cloud data centers. Virtualization and VM migration capabilities enable the data center to consolidate their computing services and use minimal number of physical servers. VM migration offers great benefits such as load balancing, server consolidation, online maintenance and proactive fault tolerance. However, in cloud computing environments the cost of VM migration requires thorough consideration. Each VM migration may result in SLA violation, hence it is essential to minimize the number of migrations to the extent possible. Failure to do so will result in performance degradation and the cloud provider will have to incur the cost in monetary terms. In previous works, the issue of SLA violation has not received thorough analysis. In this work, we devise an algorithm that will keep the migration time minimum as well as minimizing the number of migrations. This will play a major role in avoiding the performance degradation encountered by a migrating VM.

2011 IEEE Symposium on Computers and Communications (ISCC), 2011

Cloud-enabled datacenters need advanced support for an integrated management of platform virtualization technologies. The networking infrastructure of large-scale datacenters is implemented according to redundant multi-tiered architectures whose layers operate at Layer 3 of the networking stack. Splitting the network infrastructure of a datacenter in a number of IP subnets, however, creates limits to the migration of Virtual Machines, reducing the possibility for administrators to efficiently balance the load and reduce the energy consumption of the whole infrastructure. In this paper we propose an innovative solution that allows transparent migration of Virtual Machines across the whole datacenter, based on the coordinated use of NAT rules that need to be consistently managed across the layers of the datacenter networking infrastructure. We describe in details how our approach can be easily implemented with current network devices without any modification to their hardware and present an experimental evaluation of an early prototype of our solution.