Virtual Machine Deployment Research Articles

ENTICE VM Image Analysis and Optimised Fragmentation

Virtual machine (VM) images (VMIs) often share common parts of significant size as they are stored individually. Using existing de-duplication techniques for such images are non-trivial, impose serious technical challenges, and requires direct access to clouds' proprietary image storages, which is not always feasible. We propose an alternative approach to split images into shared parts, called fragments, which are stored only once. Our solution requires a reasonably small set of base images available in the cloud, and additionally only the increments will be stored without the contents of base images, providing significant storage space savings. Composite images consisting of a base image and one or more fragments are assembled on-demand at VM deployment. Our technique can be used in conjunction with practically any popular cloud solution, and the storage of fragments is independent of the proprietary image storage of the cloud provider.

GSaaS: A Service to Cloudify and Schedule GPUs

Cloud technology is an attractive infrastructure solution that provides customers with an almost unlimited on-demand computational capacity using a pay-per-use approach, and allows data centers to increase their energy and economic savings by adopting a virtualized resource sharing model. However, resources such as graphics processing units (GPUs), have not been fully adapted to this model. Although, general-purpose computing on graphics processing units (GPGPU) is becoming more and more popular, cloud providers lack of fiexibility to manage accelerators, because of the extended use of peripheral component interconnect (PCI) passthrough techniques to attach GPUs to virtual machines (VMs). For this reason, we design, develop, and evaluate a service that provides a complete management of cloudified GPUs (cGPUs) in public cloud platforms. Our solution enables an effective, anonymous, and transparent access from VMs to cGPUs that are previously scheduled and assigned by a full resource manager, taking into account newGPU selection policies and newworking modes based on the locality of the physical accelerators and the exclusivity when accessing them. This easy-to-adopt tool improves the resource availability through different cGPUs configurations for end-users, whilst cloud providers are able to achieve a better utilization of their infrastructures and offer more competitive services. Scalability results in a real cloud environment demonstrate that our solution introduces a virtually null overhead in the deployment of VMs. Besides, performance experiments reveal that GPU-enabled clusters based on cloud infrastructures can benefit from our proposal not only exploiting better the accelerators, but also serving more jobs requests per unit of time.

Energy-efficient virtual machine consolidation algorithm in cloud data centers

Cloud data centers consume a multitude of power leading to the problem of high energy consumption. In order to solve this problem, an energy-efficient virtual machine (VM) consolidation algorithm named PVDE (prediction-based VM deployment algorithm for energy efficiency) is presented. The proposed algorithm uses linear weighted method to predict the load of a host and classifies the hosts in the data center, based on the predicted host load, into four classes for the purpose of VMs migration. We also propose four types of VM selection algorithms for the purpose of determining potential VMs to be migrated. We performed extensive performance analysis of the proposed algorithms. Experimental results show that, in contrast to other energy-saving algorithms, the algorithm proposed in this work significantly reduces the energy consumption and maintains low service level agreement (SLA) violations.

Cluster-Aware Virtual Machine Collaborative Migration in Media Cloud

Media cloud has become a promising paradigm for deploying large-scale streaming media applications at a reduced cost. Due to dynamic and diverse demands of users, media cloud presents two crucial characteristics: high resource consumption and dynamic traffic among media servers. Consequently, Virtual Machine (VM) migration in media cloud is highly required to suit varying resource requirements and the dynamic traffic patterns. Moreover, migration of such bandwidth-intensive media applications in media cloud needs cautious handling, especially for the internal traffic of Data Center Networks (DCN). However, existing media cloud resource management schemes or traffic-aware VM deployment approaches are insufficient for media cloud, ignoring the characteristics of either cloud infrastructure or media streaming requirements. In this paper, we propose a cluster-aware VM collaborative migration scheme for media cloud, tightly integrating clustering, placement, and dynamic migration process. The scheme employs a clustering algorithm and a placement algorithm to obtain ideal migration strategies for newly perceived media server clusters, and a migration algorithm to effectively accomplish the migration process of media servers. Evaluation results demonstrate that our scheme can effectively migrate virtual media servers in media cloud, while reducing the total internal traffic in DCN under the resource consumption constraints of media streaming applications.

On achieving intelligent traffic-aware consolidation of virtual machines in a data center using Learning Automata

Unlike the computational mechanisms of the past many decades, that involved individual (extremely powerful) computers or clusters of machines, cloud computing (CC) is becoming increasingly pertinent and popular. Computing resources such as CPU and storage are becoming cheaper, and the servers themselves are becoming more powerful. This enables clouds to host more virtual machines (VMs). A natural consequence of this is that many modern-day data centers experience very high internal traffic within the data centers themselves. This is, of course, due to the occurrence of servers that belong to the same tenant, communicating between themselves. The problem is accentuated when the VM deployment tools are not traffic-aware. In such cases, the VMs with high mutual traffic often end up being far apart in the data center network, forcing them to communicate over unnecessarily long distances. The consequent traffic bottlenecks negatively affect both the performance of the application and the network in its entirety, posing non-trivial challenges for the administrators of these cloud-based data centers.The problem, and consequently the solution, can, quite naturally, be compartmentalized into two phases which follow each other. In the first, the task is to consolidate VMs into clusters, where those that communicate with each other fall into the same cluster. The second phase assigns these clusters onto the available server racks. Both of these phases must be executed in a traffic-aware manner. This paper provides efficient intelligent solutions for both these phases. First of all, the VMs are consolidated with a VM clustering algorithm, and this is achieved by utilizing the toolbox involving Learning Automata (LA). By mapping the clustering problem onto the Graph Partitioning (GP) problem, our LA-based solution successfully reduces the total communication cost by amounts that range between 34% and 85%. Thereafter, the resulting clusters are assigned to the server racks using a cluster placement algorithm that involves a completely different intelligent strategy, i.e., one that invokes Simulated Annealing (SA). This phase further reduces the total cost of communication by amounts that range between 89% and 99%. The analysis and results for different models and topologies demonstrate that the optimization is done in a fast and computationally-efficient way.

Minimizing SLA violation and power consumption in Cloud data centers using adaptive energy-aware algorithms

In this paper, we address the problem of reducing Cloud datacenter high energy consumption with minimal Service Level Agreement (SLA) violation. Although there are many energy-aware resource management solutions for Cloud datacenters, existing approaches focus on minimizing energy consumption while ignoring the SLA violation at the time of virtual machine (VM) deployment. Also, they do not consider the types of application running in the VMs and thus may not really reduce energy consumption with minimal SLA violation under a variety of workloads. In this paper, we propose two novel adaptive energy-aware algorithms for maximizing energy efficiency and minimizing SLA violation rate in Cloud datacenters. Unlike the existing approaches, the proposed energy-aware algorithms take into account the application types as well as the CPU and memory resources during the deployment of VMs. To study the efficacy of the proposed approaches, we performed extensive experimental analysis using real-world workload, which comes from more than a thousand PlanetLab VMs. The experimental results show that, compared with the existing energy-saving techniques, the proposed approaches can effectively decrease the energy consumption in Cloud datacenters while maintaining low SLA violation.

Explore virtual machine deployment to mobile cloud computing for multi-tenancy and energy conservation in wireless network

Mobile cloud computing (MCC) is becoming one of the primary applications and research directions of pervasive computing. Previous studies generally focuses on the following two kinds of paradigms. One is that mobile users offload computing tasks into traditional clouds to achieve collaboration processing. The other is that multiple mobile devices, as “helpers”, casually provide computing services for other mobile “requesters”. However, with the extraordinary development of mobile devices, new problems of resource utilization and multi-tenancy of mobile devices are emerging. Based on this, we propose a novel MCC paradigm combining virtual machine (VM) technologies to improve the resource utilization of mobile devices, achieve multi-tenancy and thereby obtain the more efficient MCC environment as well as a broader MCC research field. To achieve it, the VM deployment problem must be addressed firstly. The heuristic deployment approach VD-ABC based on an improved artificial bee colony algorithm is proposed. It combines with Boltzmann selection policy, to obtain the optimal solution of VM deployment. In terms of the traits of wireless network in MCC, the mapping from the obtained solution to the final VM deployment is achieved. Simulation results show that VD-ABC has a better energy saving performance while minimizing the service response latency.

Rethinking TLB Designs in Virtualized Environments

With increasing deployment of virtual machines for cloud services and server applications, memory address translation overheads in virtualized environments have received great attention. In the radix-4 type of page tables used in x86 architectures, a TLB-miss necessitates up to 24 memory references for one guest to host translation. While dedicated page walk caches and such recent enhancements eliminate many of these memory references, our measurements on the Intel Skylake processors indicate that many programs in virtualized mode of execution still spend hundreds of cycles for translations that do not hit in the TLBs. This paper presents an innovative scheme to reduce the cost of address translations by using a very large Translation Lookaside Buffer that is part of memory, the POM-TLB. In the POM-TLB, only one access is required instead of up to 24 accesses required in commonly used 2D walks with radix-4 type of page tables. Even if many of the 24 accesses may hit in the page walk caches, the aggregated cost of the many hits plus the overhead of occasional misses from page walk caches still exceeds the cost of one access to the POM-TLB. Since the POM-TLB is part of the memory space, TLB entries (as opposed to multiple page table entries) can be cached in large L2 and L3 data caches, yielding significant benefits. Through detailed evaluation running SPEC, PARSEC and graph workloads, we demonstrate that the proposed POM-TLB improves performance by approximately 10% on average. The improvement is more than 16% for 5 of the benchmarks. It is further seen that a POM-TLB of 16MB size can eliminate nearly all TLB misses in 8-core systems.

An SDN orchestrator for cloud data center: System design and experimental evaluation

AbstractThe high volatility of Virtual Machine (VM) deployments, the unpredictable traffic patterns, and the oversubscription of Data Center (DC) network links give rise to new challenges in data delivery operations in cloud DCs. These challenges can be addressed through a flexible and agile control of DC networks, which is also coordinated, i.e., orchestrated, with the management of computing resources to improve the usage of DC resources while effectively meeting user demands. In this regard, the Software‐Defined Network (SDN) paradigm can play a key role thanks to automated and multi‐granular network control operations and to a more effective interworking with applications (e.g., cloud management platforms). This paper presents an SDN‐based orchestrator able to coordinate the provision of both cloud and network services while jointly arranging VMs and delivery path allocations based on the current load. This approach assures a proper network resource utilization as well as the convey of VM data across the DC network with better than best‐effort quality. A prototype is presented with focus on the software design and on the orchestration strategies exploiting measurement‐based load estimation mechanisms. The orchestrator has been thoroughly evaluated through a comprehensive set of experiments carried out using a cloud DC testbed. Results show the effectiveness of the proposed SDN‐based orchestration approach and the impact of the orchestration strategies on the DC resource utilization.

Deployment method of VM cluster based on graph theory for cloud resource management

Cloud computing is the next generation of computation and is regarded as the fifth utility service. A core issue of cloud computing is the deployment of virtual machine (VM). A new method had been constructed for VM cluster based on graph theory in this study. The authors first described VM cluster by energy minimisation. Then, they changed deployment of VM cluster into maximum flow minimum cut problem, added sink and source point, and found activity node. Finally, activity tail-nodes were connected and cut formed for VM cluster. Experimental results show that the segmentation method can effectively realise the VM deployment clusters, can reduce the overall bandwidth requirements after deployment.

Multi-Population Ant Colony Algorithm for Virtual Machine Deployment

With the recent rapid development of cloud computing technology, how to reduce the costs of a cloud data center effectively has become an important issue. The study on virtual machine deployment mainly aims at deploying virtual machine resources required by users on a physical server rationally and effectively. This paper proposes a multi-population ant colony algorithm to solve problems of virtual machine deployment. With resource wastage and energy consumption as optimization objectives, this algorithm uses multiple ant colonies for the solution and determines strategies for information exchange among ant colonies according to the information entropy of each population to guarantee the balance of its convergence and diversity. The simulation results show that this algorithm has better performance than the single-population ant colony algorithm and can reduce resource wastage and energy consumption effectively for high-demand virtual machine deployment.

Multiobjective Level-Wise Scientific Workflow Optimization in IaaS Public Cloud Environment

Cloud computing in the field of scientific applications such as scientific big data processing and big data analytics has become popular because of its service oriented model that provides a pool of abstracted, virtualized, dynamically scalable computing resources and services on demand over the Internet. However, resource selection to make the right choice of instances for a certain application of interest is a challenging problem for researchers. In addition, providing services with optimal performance at the lowest financial resource deployment cost based on users’ resource selection is quite challenging for cloud service providers. Consequently, it is necessary to develop an optimization system that can provide benefits to both users and service providers. In this paper, we conduct scientific workflow optimization on three perspectives: makespan minimization, virtual machine deployment cost minimization, and virtual machine failure minimization in the cloud infrastructure in a level-wise manner. Further, balanced task assignment to the virtual machine instances at each level of the workflow is also considered. Finally, system efficiency verification is conducted through evaluation of the results with different multiobjective optimization algorithms such as SPEA2 and NSGA-II.

Cloud and network orchestration in SDN data centers: Design principles and performance evaluation

The oversubscription of Data Center network links and the high volatility of Virtual Machine (VM) deployments call for a flexible and agile control of Data Center networks, coordinated with computing resource control (i.e., cloud resource management). The Software-Defined Network (SDN) paradigm opens up new opportunities to design convergent resource management systems able to address the provisioning of cloud services while meeting dynamically changing traffic demands of running VMs.This paper presents the architectural design of an SDN-based orchestration system which is able to coordinate the provision of composite cloud and network services while assuring computational requirements as well as a better than best effort VM data delivery. The proposed orchestration system is able to perform VM allocations also based on estimations of switch/link and server loads as result of the synergistic interwork of the following functions: (i) resource selection and composition functions, (ii) coordinated resource configuration and management functions, (iii) monitoring and registration functions of resource status. A set of resource selection and composition strategies and estimation schemes have been also specified.The orchestration process has been thoroughly evaluated through a comprehensive set of simulations that clearly show an increasing acceptance rate of service requests, an improved utilization of network capabilities while effectively preventing significant degradations of the user experience despite the oversubscription of data center network links.

A Novel Method for Virtual Machine Placement Based on Euclidean Distance

With the increasing popularization of cloud computing, how to reduce physical energy consumption and increase resource utilization while maintaining system performance has become a research hotspot of virtual machine deployment in cloud platform. Although some related researches have been reported to solve this problem, most of them used the traditional heuristic algorithm based on greedy algorithm and only considered effect of single-dimensional resource (CPU or Memory) on energy consumption. With considerations to multi-dimensional resource utilization, this paper analyzed impact of multi-dimensional resources on energy consumption of cloud computation. A multi-dimensional resource constraint that could maintain normal system operation was proposed. Later, a novel virtual machine deployment method (NVMDM) based on improved particle swarm optimization (IPSO) and Euclidean distance was put forward. It deals with problems like how to generate the initial particle swarm through the improved first-fit algorithm based on resource constraint (IFFABRC), how to define measure standard of credibility of individual and global optimal solutions of particles by combining with Bayesian transform, and how to define fitness function of particle swarm according to the multi-dimensional resource constraint relationship. The proposed NVMDM was proved superior to existing heuristic algorithm in developing performances of physical machines. It could improve utilization of CPU, memory, disk and bandwidth effectively and control task execution time of users within the range of resource constraint.

Handling Boot Storms in Virtualized Data Centers—A Survey

Large-scale virtual machine (VM) deployment in virtualized data centers is a very slow process. This is primarily due to the resource bottlenecks that are created at the storage, network, and host physical machines when a large number of VMs are requested simultaneously. For companies that provide virtual desktops to their employees, it is common to encounter such requests each day, when their employees turn up for work. In addition, a large number of VMs are often required to be deployed instantly, in order to absorb a spike in the workload, at online e-commerce websites. In such scenarios, long deployment times are unacceptable, and reducing them is of paramount importance. In this article, we first abstract out the key techniques suggested in the literature to speed up this deployment process. We follow this with a classification of these techniques into a taxonomy and propose a framework that can be used to compare them. Finally, we identify problem areas that warrant further research and bring out the shortcomings of the current state-of-the-art solutions.

Resource pre-allocation algorithms for low-energy task scheduling of cloud computing

In order to lower the power consumption and improve the coefficient of resource utilization of current cloud computing systems, this paper proposes two resource pre-allocation algorithms based on the shut down the redundant, turn on the demanded strategy here. Firstly, a green cloud computing model is presented, abstracting the task scheduling problem to the virtual machine deployment issue with the virtualization technology. Secondly, the future workloads of system need to be predicted: a cubic exponential smoothing algorithm based on the conservative control (CESCC) strategy is proposed, combining with the current state and resource distribution of system, in order to calculate the demand of resources for the next period of task requests. Then, a multi-objective constrained optimization model of power consumption and a low-energy resource allocation algorithm based on probabilistic matching (RA-PM) are proposed. In order to reduce the power consumption further, the resource allocation algorithm based on the improved simulated annealing (RA-ISA) is designed with the improved simulated annealing algorithm. Experimental results show that the prediction and conservative control strategy make resource pre-allocation catch up with demands, and improve the efficiency of real-time response and the stability of the system. Both RA-PM and RA-ISA can activate fewer hosts, achieve better load balance among the set of high applicable hosts, maximize the utilization of resources, and greatly reduce the power consumption of cloud computing systems.

<italic>EnReal</italic>: An Energy-Aware Resource Allocation Method for Scientific Workflow Executions in Cloud Environment

Scientific workflows are often deployed across multiple cloud computing platforms due to their large-scale characteristic. This can be technically achieved by expanding a cloud platform. However, it is still a challenge to conduct scientific workflow executions in an energy-aware fashion across cloud platforms or even inside a cloud platform, since the cloud platform expansion will make the energy consumption a big concern. In this paper, we propose an Energy-aware Resource Allocation method, named EnReal, to address the above challenge. Basically, we leverage the dynamic deployment of virtual machines for scientific workflow executions. Specifically, an energy consumption model is presented for applications deployed across cloud computing platforms, and a corresponding energy-aware resource allocation algorithm is proposed for virtual machine scheduling to accomplish scientific workflow executions. Experimental evaluation demonstrates that the proposed method is both effective and efficient.

New Deadline-Aware Energy-Consumption Optimization Model and Genetic Algorithm Under Cloud Computing

With large-scale data centers widely deployed around the world, their huge energy consumption becomes a primary concern. Effective resource allocation and scheduling is one of the key to solve this problem. However, existing studies on this topic are relatively rare. In this paper, a new deadline-aware energy-consumption optimization model is designed, which optimizes both the idle and execution energy consumption of servers. To save the idle energy consumption, we propose a new virtual machine deployment algorithm for mapping virtual machines to a constrained packing problem with multidimensional variables. In the proposed genetic algorithm, in order to improve the diversity of the population, we select some of the individuals which do not satisfy time constraints but have low energy consumption into the next generation. To save the execution energy consumption, we adopt the technique of dynamic voltage and frequency scaling. Finally, experimental results show that compared with the existing algorithms, the proposed one greatly reduces the total energy consumption of data centers under the time constraints of tasks.

A Two-Tier Energy-Aware Resource Management for Virtualized Cloud Computing System

The economic costs caused by electric power take the most significant part in total cost of data center; thus energy conservation is an important issue in cloud computing system. One well-known technique to reduce the energy consumption is the consolidation of Virtual Machines (VMs). However, it may lose some performance points on energy saving and the Quality of Service (QoS) for dynamic workloads. Fortunately, Dynamic Frequency and Voltage Scaling (DVFS) is an efficient technique to save energy in dynamic environment. In this paper, combined with the DVFS technology, we propose a cooperative two-tier energy-aware management method including local DVFS control and global VM deployment. The DVFS controller adjusts the frequencies of homogenous processors in each server at run-time based on the practical energy prediction. On the other hand, Global Scheduler assigns VMs onto the designate servers based on the cooperation with the local DVFS controller. The final evaluation results demonstrate the effectiveness of our two-tier method in energy saving.

Joint study on VMs deployment, assignment and migration in geographically distributed data centers

Enterprises build private clouds to provide IT resources for geographically distributed subsidiaries or product divisions. Public cloud providers like Amazon lease their platforms to enterprise users, thus, enterprises can also rent a number of virtual machines (VMs) from their data centers in the service provider networks. Unfortunately, the network cannot always guarantee stable connectivity for their clients to access the VMs or low-latency transfer among data centers. Usually, both latency and bandwidth are in unstable network environment. Being affected by background traffics, the network status can be volatile. To reduce the latency uncertainty of client accesses, enterprises should consider the network status when they deploy data centers or rent virtual data centers from cloud providers. In this paper, we first develop a data center deployment and assignment scheme for an enterprise to meet its users' requirements under uncertain network status. To accommodate to the changes of the network status and users' demands, a VMs migration-based redeployment scheme is adopted. These two schemes work in a joint way, and lay out a framework to help enterprises make better use of private or public clouds.