Performance analysis of local, network and distributed file systems running inside user’s virtual machines in cloud environment

In this paper, a research work is carried out to improve workload management by minimize energy consumption of node, localizing traffic in a network and reduce communication delay between virtual machines in cloud environment. Cloud computing consists numerous virtual machine to store huge volume of data. The usage of virtual machines is to provide separate guest operating system which is popular now-a-days. The workload management across different virtual machines in cloud is really a difficult process. Decentralized approaches were proposed in the literature for energy efficient management of virtual machines. But still it has some issues like failed to consider the communication between virtual machines and it leads to traffic in a network. Thus it reduces the performance of work load management.

Nowadays, review mining is very popular. People usually want to go for the reviews first rather than directly going to the market and purchase the product or paying for a service. However, it is very difficult to get the best one until making the comparison between two or more positively categorized products. This work highlights a platform where review mining become very easy and automated. The automation of crawling is done through various websites for a product or service and downloading their reviews at their own. The model is deployed in the Cloud environment for executing the multiple websites in parallel, which speed up the process to meet the requirements of real time recommendations. Experiments are performed on various product reviews datasets in order to provide and analyze the reviews in the form of star rating score scaling from 0–5.

The virtual machine placement problem can be described as designing optimal placement scheme for virtual machine in cloud environment. Cloud data centers are facing increasingly virtual machine placement problems, such as high energy consumption, imbalanced utilization of multidimensional resource, and high resource wastage rate. In this paper, typical exact and heuristic algorithms as solution to the virtual machine placement problem in the cloud are surveyed in terms of energy consumption and resource wastage. The purpose of this paper is to evaluate the performance of both the exact and approximate algorithms developed by using the WebCloudSim sytem.

In recent decades, task scheduling and load balancing in the cloud is a growing research area, due to the vast amount of data stored in the server highly increases the load. In order to address this concern, Hybrid Max-Min Genetic Algorithm (HMMGA) is proposed for task scheduling and load balancing in the cloud environment. At first, the load is evaluated for every Virtual Machine (VM), if the load is high, then HMMGA is used for balancing the load. HMMGA selects the best VMs to assign the tasks and migrates the over-loaded VMs tasks to the under-loaded VMs. HMMGA significantly avoids the imbalanced workload performance in the cloud environment. In this research paper, the proposed HMMGA performance is compared to Max-Min algorithm, Low time complexity and low cost binary Particle Swarm Optimizer (IBPSO-LBS) and PSO with Technique of Order Preference by Similarity to Ideal Solution (TOPSIS) algorithm to examine the efficacy of HMMGA. From the experimental simulation, the result shows that HMMGA averagely delivers 1.63 and 3.88 seconds less make span compared to the Max-Min and TOPSIS-PSO algorithm for five VMs. In addition, HMMGA averagely enhances 10% to 40% of resource utilization than the MaxMin and TOPSIS-PSO algorithm. In another experiment, the HMMGA approximately showed 1.7 to 25.99 seconds less average waiting time compared to the Max-Min and IBPSO-LBS.

ABSTRACT: An agent-less technique to understand virtual machines (VMs) behavior and their changes during the VM life-cycle is essential for many performance analysis and debugging tasks in the cloud environment. Because of privacy and security issues, ease of deployment and execution overhead, the method preferably limits its data collection to the physical host level, without internal access to the VMs. We propose a host-based, precise method to recover execution flow of virtualized environments, regardless of the level of virtualization. Given a VM, the Any-Level VM Detection Algorithm (ADA) and Nested VM State Detection (NSD) Algorithm compute its execution path along with the state of virtual CPUs (vCPUs) from the host kernel trace. The state of vCPUs is displayed in an interactive trace viewer (TraceCompass) for further inspection. Then, a new approach for profiling threads and processes inside the VMs is proposed. Our proposed VM trace analysis algorithms have been open-sourced for further enhancements and to the benefit of other developers. Our new techniques are being evaluated with workloads generated by different benchmarking tools. These approaches are based on host hypervisor tracing, which brings a lower overhead (around 1%) as compared to other approaches.

The rapid growth in cloud computing encourages the cloud provider to create more reliable, scalable and efficient cloud environment. For providing such a scalable environment, cloud provider develops large-scale Cloud Data Centers (CDC). Such CDCs consume a large amount of electrical energy for operational purpose and cooling purpose to keep the machines operating at a right temperature, but it also emits a large amount of carbon-dioxide. Thus to decrease the energy consumption, Virtual Machines (VMs) are consolidated to fewer numbers of active servers through VM migration techniques, this leads to Service Level Agreement (SLA) violation and performance degradation. Therefore to retain the energy-performance tradeoff, we should perform an optimum number of VM migrations. There are two cases when the VM migration performed: First when a hot spot is found and Second when a cold spot is found. But in both the cases we do not directly migrate the VM instead first we check whether it is required to migrate the VM. After that, the algorithm finds an appropriate destination host on which the VM will be migrated. In this paper, we implement three approaches to select destination host: First-fit, Best-fit, and Worst-fit. Finally compare the result of a number of VM migration performed, energy consumption and SLA violations with existing one.

This paper addresses the impact of Virtual Memory Streaming (VMS) technique in provisioning virtual machines (VMs) in cloud environment. VMS is a scaling virtualization technology that allows different virtual machines rapid scale, high performance, and increase hardware utilization. Traditional hypervisors do not support true no-downtime live migration, and its lack of memory oversubscription can hurt the economics of a private cloud deployment by limiting the number of VMs on each host. VMS brings together several advanced hypervisor memory management techniques including granular page sharing, dynamic memory footprint management, live migration, read caching, and a unique virtual machine cloning capability. An architecture model is described, together with a proof-of-concept implementation, that VMS dynamically scaling of virtualized infrastructure with true live migration and cloning of VMs. This paper argues that VMS for Cloud allows requiring significantly reduced server memory and reducing the time for virtualized resource scaling by instantly adding more virtual machines.

Due to the high growth of SNS population, service scalability is one of the critical issues to be addressed. The cloud environment provides the flexible computing and storage resources for services deployment, which fits the characteristics of scalable SNS deployment. However, if the SNS related information is not properly placed, it will cause unbalance load and heavy transmission cost on the storage virtual machine (VM) and cloud data center (CDC) network. In this paper, we characterize the SNS into a graph model based on the users associations and interest correlations. The node weight represents the degree of associations, which can be indexed by the number of friends or data sources, and the link weight denotes the correlation between users/data sources. Then, based on the SNS graph, the two-step algorithm is proposed in this paper to determine the placement of SNS related data among VMs. Two k-means based clustering schemes are proposed to allocate social data in proper VM and physical servers for pre-configured VM and dynamic VM environment, respectively. The experimental example was conducted and to illustrate and compare the performance of the proposed schemes.

In cloud computing environment, the load balancing of non pre-emptive independent tasks is an important aspect of task scheduling. The tasks are executed on Virtual Machines (VMs) and these VMs are run in parallel so that the load has to be well balanced across all VMs. Since the cloud user pay for the service he/she is interested in reducing the execution time of task as well as the cost of using VM instances (CPU, Memory, Bandwidth etc). The proposed algorithm balance the load across VMs effectively by mapping tasks to under loaded or idle VMs based on the foraging behaviour of honey bees and if there is more than one under loaded VMs, it also selects cost efficient one using Pareto dominance strategy. The experimental results show that the algorithm is effective when compared with existing algorithms and it also reduces the cost of using VM instances.

Cloud computing is distributed computing, storing, sharing and accessing data over the Internet. It provides a pool of shared resources to the users available on the basis of pay as you go service that means users pay only for those services which are used by him according to their access times. Load balancing ensures that no single node will be overloaded and used to distribute workload among multiple nodes. It helps to improve system performance and proper utilization of resources. We propose an improved load balancing algorithm for job scheduling in the cloud environment using K-Means clustering of cloudlets and virtual machines in the cloud environment. All the cloudlets given by the user are divided into 3 clusters depending upon client’s priority, cost and instruction length of the cloudlet. The virtual machines inside the datacenter hosts are also grouped into multiple clusters depending upon virtual machine capacity in terms of processor, memory, and bandwidth. Sorting is applied at both the ends to reduce the latency. Multiple number of experiments have been conducted by taking different configurations of cloudlets and virtual machine. Various parameters like waiting time, execution time, turnaround time and the usage cost have been computed inside the cloudsim environment to demonstrate the results. Compared with the other job scheduling algorithms, the improved load balancing algorithm can outperform them according to the experimental results.Â

Multiple instances of virtual machines can run on a single physical host sharing hardware and software resources in cloud computing. One of the resources that is shared among multiple Virtual Machines (VM) in the cloud is Cache. Such Virtual machines are targeted for an abnormal activity like side channel attack. Cache-based side channel attack is one of the side channel attack in cloud environment which leaks the private information of the client. The proposed approach includes the detection and mitigation of cache-based side channel attack in cloud infrastructure. The proposed approach comprises of three components: a collection of virtual machine status, the Fuzzy controller to detect attack and mitigation. The fuzzy rule-based controller is incorporated in this approach to identify the cache-attack on the log file. This system works dynamically to prevent cache attacks on the cloud environment and will incur very small overhead in performance.

Optimization of data intensive applications is affected greatly by the nature of the platform, the distributed file system, the co-location of data and programs, and the proximity of resources. Data intensive applications running on a public cloud have been shown to exhibit degraded performance compared to a private cluster. This performance degradation is as a result of the inefficient resource allocation adopted in a virtualized, volatile and multi-tenancy environment such as a cloud. The placement of virtual machines is critical for improving performance in geo-clouds environment. We address degradation in performance by designing a scheduling algorithm that dynamically distributes virtual machines based on the characteristics of the job, the network characteristics and the real time performance of the data centers. Our algorithm also finds the best node(s) to host virtual machines that will yield maximum resource utilization and minimize bandwidth consumption. Our results show that incorporating these characteristics in a resource scheduling algorithm increases the performance of data intensive application than known traditional scheduling.

With the expansion of the cloud computing usage over a wide range of areas and different kinds of users, the cloud providers are taking full advantage of all their resources as much as they can. Memory is the most expensive resource in terms of oversubscribing and this has resulted in high price to the end user. Furthermore, performing swapping in Virtual Machines (VM) is expensive, so the cloud provider usually do not offer any swapping space for its systems. As a consequence, when a VM runs out of memory, user processes are killed. This scenario in the cloud environment is especially critical, since the user loses all of his/her execution time and, by extension, the money invested in this computation. This paper addresses this critical problem by providing a kernel extension that monitors the memory requirements of a VM and prevents the out of memory state by creating swapping space dynamically. The paper describes the design and implementation of a preliminary prototype of this kernel extensions and evaluates its performance.

Live virtual machine (VM) migration is a technique for achieving system load balancing in a cloud environment by transferring an active VM from one physical host to another. This technique has been proposed to reduce the downtime for migrating overloaded VMs, but it is still time- and cost-consuming, and a large amount of memory is involved in the migration process. To overcome these drawbacks, we propose a Task-based System Load Balancing method using Particle Swarm Optimization (TBSLB-PSO) that achieves system load balancing by only transferring extra tasks from an overloaded VM instead of migrating the entire overloaded VM. We also design an optimization model to migrate these extra tasks to the new host VMs by applying Particle Swarm Optimization (PSO). To evaluate the proposed method, we extend the cloud simulator (Cloudsim) package and use PSO as its task scheduling model. The simulation results show that the proposed TBSLB-PSO method significantly reduces the time taken for the load balancing process compared to traditional load balancing approaches. Furthermore, in our proposed approach the overloaded VMs will not be paused during the migration process, and there is no need to use the VM pre-copy process. Therefore, the TBSLB-PSO method will eliminate VM downtime and the risk of losing the last activity performed by a customer, and will increase the Quality of Service experienced by cloud customers.

We introduce a new scheduling method that reduces the time required to deploy applications by parallelizing and optimizing a schedule on multiple virtual machines (VMs) in cloud computing. Enterprise solutions typically consist of multiple servers and clients with complicated installation dependencies. If users deploy such a solution, they need to install the software sequentially in the proper installation order without violating the dependencies. Auto-deployment tools such as Chef and Puppet take a long time to install multiple applications on multiple VMs in sequential order because these tools cannot find the dependencies automatically and cannot optimize the deployment schedules. Users need to configure and schedule the deployment carefully after they understand the entire architecture of the enterprise solution. Our new automated deployment approach, Parallelized and Optimized Deployment Scheduling (PODS), can detect these application dependencies and optimize the deployment schedules based on two profiles: 1) the network connectivity between a server and clients, and 2) the time required to install applications for sequential deployment. PODS optimization can reduce the duplication timings of multiple VMs' requests to download applications from the same repository server in the parallelized deployment, which reduces network congestion of the repository server. We implemented PODS with Chef and evaluated it on a commercial public cloud environment. PODS reduced the time to deploy four VMs by 36% compared with the standard sequential deployment method.