A Scalable Virtualized Server Cluster Providing Sensor Data Storage and Web Services

Virtualization has been shown to be an attractive path to increase overall system resource utilization. The use of live virtual machine (VM) migration has enabled more effective sharing of system resources across multiple physical servers, resulting in an increase in overall performance. Live VM migration can be used to load balance virtualized clusters. To drive live migration, we need to be able to measure the current load imbalance. Further, we also need to accurately predict the resulting load imbalance produced by any migration.In this paper we present a new metric that captures the load of the physical servers and is a function of the resident VMs. This metric will be used to measure load imbalance and construct a load-balancing VM migration framework. The algorithm for balancing the load of virtualized enterprise servers follows a greedy approach, inductively predicting which VM migration will yield the greatest improvement of the imbalance metric in a particular step. We compare our algorithm to the leading commercially available load balancing solution - VMware's Distributed Resource Scheduler (DRS). Our results show that when we are able to accurately measure system imbalance, we can also predict future system state. We find that we can outperform DRS and improve performance up to 5%. Our results show that our approach does not impose additional performance impact and is comparable to the virtual machine monitor overhead.

Linux Virtual Server (LVS) cluster technology is commonly used to provide highperformance web service. In LVS cluster, WLC scheduling algorithm is often adopted and has good effect. Based on the WLC scheduling algorithm, this paper presents a kind of simple optimization method by using server's performance indexes to calculate server's weight. The performance of the LVS cluster can be effectively improved by the optimized method.

Cloud computing models use virtual machine (VM) clusters for protecting resources from failure with backup capability. Cloud user tasks are scheduled by selecting suitable resources for executing the task in the VM cluster. Existing VM clustering processes suffer from issues like preconfiguration, downtime, complex backup process, and disaster management. VM infrastructure provides the high availability resources with dynamic and on-demand configuration. The proposed methodology supports VM clustering process to place and allocate VM based on the requesting task size with bandwidth level to enhance the efficiency and availability. The proposed clustering process is classified as preclustering and postclustering based on the migration. Task and bandwidth classification process classifies tasks with adequate bandwidth for execution in a VM cluster. The mapping of bandwidth to VM is done based on the availability of the VM in the cluster. The VM clustering process uses different performance parameters like lifetime of VM, utilization of VM, bucket size, and task execution time. The main objective of the proposed VM clustering is that it maps the task with suitable VM with bandwidth for achieving high availability and reliability. It reduces task execution and allocated time when compared to existing algorithms.

In this paper, we consider virtual machine (VM) placement problem for cloud data centers. Different from previous works, the VMs considered need to be conducted within a finite duration. However, due to the limited resource at physical machines (PMs), the VM may be unable to place on a single PM within duration. To make sure the successful placement, the VM may undergo several PMs for service, e.g., it is run on one PM at this time, while migrated to another PM next time. Migrating the VM from one PM to another incurs extra cost, e.g., the bandwidth. In this paper, we formulate a joint optimization problem, which maximizes the revenue of successfully placing the VMs, and minimizes the total number of migrations for the VMs. We prove it is NP-hard, and show that the minimum number of migrations for each VM under the resource constraint of PMs is equivalent to finding a shortest path problem. We then design a heuristic algorithm, which considers both the revenue and the minimum number of migrations required for the VM. We also design a 1 over 2-approximation algorithm for a special case of the problem. Finally, we demonstrate the effectiveness of the proposed scheme through simulations.

Nowadays cloud computing provides an effective way of implementing infrastructure as a service (IaaS). However virtualized data centers still face many challenges, such as low resource utilization of physical machines (PMs) and imbalanced server loads. Virtual machine (VM) consolidation based on live migration allows administrator to dynamically redeploy VMs into PMs for better resource utilization. Common VM consolidation methods usually focus on one challenge, and pay little attention to others or just ignore them, while effective VM redeployment should make tradeoffs between these challenges, and more importantly, should not let other challenges become worse. On the other hand, since VM live migration leads to performance degradation of applications, consolidation work should control migration cost. In this paper, we provide a manner to comprehensively consider power consumption, load balancing, communication delay and migration cost during VM redeployment. And we formalize VM consolidation as a multiobjective optimization problem, then solve this problem with an improved genetic algorithm. Simulation experiments based on real world workload trace show that compared with single objective optimization approaches our method effectively make tradeoffs between optimized objectives and has better overall performance, which is more practical in real data centers.

Evolution of virtualization in data center is fast and is being widely used in the world. Usage of virtual machine (VM) in data center (DC) cannot be aside from problems such as: Operating System (OS) problem, virtual network, memory and CPU utilizations. Besides of those problems, utilization of VM can be found in cloud computing technology to make more efficient and its performance should be good entirely. Problem in VM is very complex. It can be found in OS, application and physical server. To keep it in good performance, operational engineer should operate VM monitoring system in order to keep VM run well. This research will use several methods, such as: fuzzy Mamdani, holdout validation and naive Bayes. These methods will then create decision making for VM performance condition.

Dealing with Small Files Problem in Hadoop Distributed File System

This paper proposes to add the multithreaded Graphic Processing Units (GPUs) to some virtual machines (VMs) in the existing cloud-based VM groups. To handle the multidimensional or multithreaded computing that a CPU cannot process quickly by a GPU that has hundreds of Arithmetic Logic Units (ALUs), and to regulate the time for initiating physical servers by real-time thermal migration, our proposed scheme can enhance the system performance and reduce the energy consumption of long-term computing. Four major techniques in this paper include: (1) GPU virtualization, (2) Hypervisor for GPU, (3) Thermal migration implementation, and (4) Estimation of multithreaded tasks. In no matter quantum mechanics, astronomy, fluid mechanics, or atmospheric simulation and prediction, a GPU suits not only parallel multithreaded computing for its tens of times performance than a CPU, but also multidimensional array operations for its excellent efficiency. Therefore, how to distribute the computing performance of CPUs and GPUs appropriately becomes a significant issue. In general cloud computing applications, it is rarely seen that GPUs can outperform CPUs. Furthermore, for groups of virtual servers, many tasks actually can be completed by CPUs without the support of GPUs. Thus, it is a waste of resources to implement GPUs to all physical servers. For this reason, by integrating with the migration characteristic of VMs, our proposed scheme can estimate whether to compute tasks by physical machines with GPUs or not. In estimating tasks, we use Amdahl’s law to estimate the overall performance include communication delays, Synchronization overhead and me possible additional burden.KeywordsVirtual Machine (VM)MultithreadingGPUCUDAMapreduce

Optimizing dynamic survivability and security of replicated data in cloud systems under co-residence attacks

Web services have increasingly begun to rely on public cloud platforms. The virtualization technologies employed by public clouds can however trigger contention between virtual machines (VMs) for shared physical machine (PM) resources thereby leading to performance problems for the Web service. Past studies have exploited PM level performance metrics such as Clock Cycles Per Instruction to detect such platform induced performance interference. Unfortunately, public cloud customers do not have access to such metrics. They can typically only access VM-level metrics and application level metrics such as transaction response times and such metrics alone are often not useful for detecting inter-VM contention. This poses a difficult challenge to Web service operators for detecting and managing platform induced performance interference issues inside the cloud. We propose a machine learning based interference detection technique to address this problem. The technique applies collaborative filtering to predict whether a given transaction being processed by a Web service is suffering adversely from interference. The results can then be used by a management controller to trigger remedial actions, e.g., reporting problems to the system manager or switching cloud providers. Results using a realistic Web benchmark show that the approach is effective. The most effective variant of our approach is able to detect about 96% of performance interference events with almost no false alarms.

Virtual machine cluster mobility in inter-cloud platforms

Cloud computing is a very well known technology for all business people, software developers, end-users, and so on. Significant researches are going on to balance the cloud load. The migration of heavily loaded Virtual Machines (VMs) into lightly loaded Physical Machines (PMs) balances the Cloud load. In Resource Intensity Aware Load Balancing (RIAL) method, based on the weight of resources under utilization, it selected the VMs from heavily loaded PMs for migration and chosen the lightly loaded PMs as destination. An Improved RIAL was proposed to consider both lightly and heavily loaded PMs as destination. Later it was enhanced in the proposed Power Consumption Aware- Traffic Aware- IRIAL (PT-IRIAL) method with the consideration of power consumption, temperature and traffic measures to select the VMs for migration and select PMs for destination. From all these, in this current paper, the crossover and mutation process of GA is utilized to optimally select the migration VMs and choose the destination PMs. Thus this GA based load optimization algorithm optimally maps the migration VMs with the destination PMs efficiently.

Chapter 7 - Virtual Disks

Virtualization technologies are extensively utilized in data centers, particularly cloud computing. This facilitates data center management and diminishes the number of physical machines (servers) and, subsequently, their cooling requirements, leading to cost, space, and power consumption reductions. When applications in data centers are executing independent parallel transactions, but with similar performance requirements, the appropriate level of virtual machine consolidation on a server poses a fundamental challenge for capacity planning. This article introduces a method to evaluate the performance speedup achieved through virtualization on any server and the effects of virtualization and consolidation overheads on physical or virtual machine scalability. This research formalizes the speedup and overheads, using classical computer architecture statements. but at the same time proposes a new method to analyze these overhead amounts and types, showing the scalability and efficiency of different consolidations in the same server and its comparison against no consolidation. This work also proposes a new way to determine the optimal number of physical servers and the optimal number of consolidated virtual machines for a given transaction workload. The real experimentation was performed with different workload sizes, types of virtualizations and different servers. The method presented also facilitates the representation of linear scalability against the real degree of parallelism of either physical machines or consolidated virtual machines for a given transaction workload, as well as striking the right balance between speedup and energy in virtual server consolidation.

Infrastructure-as-a-Service (IaaS) clouds have become more popular enabling users to run applications under virtual machines. This paper investigates the energy-aware virtual machine (VM) allocation problems in IaaS clouds along characteristics: multiple resources, and fixed interval times and non-preemption of virtual machines. Many previous works proposed to use a minimum number of physical machines, however, this is not necessarily a good solution to minimize total energy consumption in the VM placement with multiple resources, fixed interval times and non-preemption. We observed that minimizing total energy consumption of physical machines is equivalent to minimize the sum of total completion time of all physical machines. Based on the observation, we propose EMinRET algorithm. The EMinRET algorithm swaps an allocating VM with a suitable overlapped VM, which is of the same VM type and is allocated on the same physical machine, to minimize total completion time of all physical machines. The EMinRET uses resource utilization during executing time period of a physical machine as the evaluation metric, and will then choose a host that minimizes the metric to allocate a new VM. In addition, this work studies some heuristics for sorting the list of virtual machines (e.g., sorting by the earliest starting time, or the longest duration time first, etc.) to allocate VM. Using the realistic log-trace in the Feitelson's Parallel Workloads Archive, our simulation results show that the EMinRET algorithm could reduce from 25% to 45% energy consumption compared with power-aware best-fit decreasing (PABFD) [1]) and vector binpacking norm-based greedy algorithms (VBP-Norm-L1/L2 [2]). Moreover, the EMinRET heuristic has also less total energy consumption than our previous heuristics (e.g. MinDFT and EPOBF) in the simulations (using same virtual machines sorting method).