Delay-aware network I/O virtualization for cloud computing

While modern datacenters are increasingly adopting virtual machines (VMs) to provide elastic cloud services, they still rely on traditional TCP for congestion control. In virtualized datacenters, TCP endpoints are separated by a virtualization layer and subject to the intervention of the hypervisor's scheduling. Most previous attempts focused on tuning the hypervisor layer to try to improve the VMs' I/O performance, and there is very little work on how a VM's guest OS may help the transport layer to adapt to the virtualized environment. In this paper, we find that VM scheduling delays can heavily contaminate RTTs as sensed by VM senders, preventing TCP from correctly learning the physical network condition. After giving an account of the source of the problem, we propose PVTCP, a ParaVirtualized TCP to counter the distorted congestion information caused by VM scheduling on the sender side. PVTCP is self-contained, requiring no modification to the hypervisor. Experiments show that PVTCP is much more effective in addressing incast congestion in virtualized datacenters than standard TCP.

Since a virtual machine independently uses its own caching policy, redundant disk operations exacerbate the I/O virtualization overhead when virtual machines access large amounts of data on shared storage. This paper presents XHive, an efficient cooperative caching system that is implemented at the virtualization layer, for consolidated environments. Our proposed scheme globally manages buffer caches of consolidated virtual machines in order to accommodate a shared working set in machine memory. A singlet, which is a block cached solely by a virtual machine, is preferentially given more chances to be cached in machine memory by XHive, when it is evicted by a guest operating system. For efficient use of limited memory, singlets are cached in memory that is collaboratively donated from idle memory of virtual machines. Our evaluation shows that XHive significantly reduces disk I/O operations for shared working sets, thereby achieving high read performance and scalability. Improved scalability enables a high degree of workload consolidation with respect to virtual machines that have shared working sets.

Hypervisor-enabled virtualization is a promising technology to deploy applications and implement Virtual Machine (VM) migration on cloud data centers. However, one critical aspect in hypervisor-based solution is that to deploy applications each VM is installed with a dedicated guest OS along with the requirement of binaries and library files and substantially large VM's image size. Therefore, it is a significant concern for data center administrators because it causes performance overheads. To solve these issues, a new paradigm of virtualization technology that has attracted considerable attention is containerization. Although, progress has been made to address container-based virtualization, there has been less attention to explore the issue of performance evaluation of container and VM in a migration environment. The main contribution of this research study is to develop an experimental setup to implement and evaluate the performance of LXD/CR container-based migration technique and compared it with the pre-copy VM migration scheme. To conduct this performance evaluation, first we modified Linux kernel and implemented LXD which uses liblxc to launch container, then migration of the running container is facilitated using a checkpoint/restore mechanism of CRIU technique. To ensure the applicability and validity for real-time cloud infrastructure, we executed a wide variety of benchmarks such as web server, CPU-intensive, disk I/O and database. From the results obtained, it is found that compared to pre-copy VM migration scheme, LXD/CR container migration technique reduces the downtime, migration time, amount of data transferred and the number of pages transferred by 75.66%, 65.55%, 76.63%, and 76.78%, respectively. In addition, the migration overhead caused is reduced by 55.89% w.r.t. CPU utilization and 76.52% w.r.t. RAM utilization. This being an important contribution for migrating the running applications efficiently and consequently increase the business value of the cloud infrastructure providers.

Cloud computing is a new computation technology that provides services to consumers and businesses. The main idea of Cloud computing is to present software and hardware services through the Internet to the users and organizations at all levels. In Cloud computing, the users pay for the services, which means a usage-based payment system is used in this technology. Using virtualization technology in computation resources enables the appropriate utilization of resources in cloud computing. One of the most significant challenging issues in virtualization technology is the placement of optimal virtual machines on physical machines in cloud data centers. The placement of virtual machines comprises a process wherein virtual machines are mapped onto physical machines in cloud data centers. Optimal deployment leads to the reduction in power consumption, optimal use of resources, traffic reduction in data centers, costs reduction, and efficiency enhancement of the data center in the cloud. The present article proposed a new approach using a combination of the Sine–Cosine Algorithm and Salp Swarm Algorithm as discrete multi-objective and chaotic functions for optimal virtual machine placement. The first goal of the proposed algorithm was to reduce the power consumption in cloud data centers by condensing the number of active physical machines. The second goal was to reduce the waste of resources and manage it by optimally virtual machine placement on physical machines in cloud data centers. The third objective was to minimize and reduce Service Level Agreement among the active physical machines in cloud data centers. The proposed method prevent the increase in the migration of virtual machines onto physical machines. Ultimately, the results obtained from the proposed algorithm were compared with those of previous akin algorithms in the literature, including First Fit, Virtual Machine Placement Ant Colony System, and Modified Best Fit Decreasing. The proposed scheme is tested using Amazon EC2 Instances and the result indicated that the proposed algorithm performs better than the existing algorithms for various performance metrics.

Virtual Machine (VM) allocation for multiple tenants is an important and challenging problem to provide efficient infrastructure services in cloud data centers. Tenants run applications on their allocated VMs, and the network distance between a tenant's VMs may considerably impact the tenant's Quality of Service (QoS). In this study, we define and formulate the multi-tenant VM allocation problem in cloud data centers, considering the VM requirements of different tenants, and introducing the allocation goal of minimizing the sum of the VMs' network diameters of all tenants. Then, we propose a Layered Progressive resource allocation algorithm for multi-tenant cloud data centers based on the Multiple Knapsack Problem (LP-MKP). The LP-MKP algorithm uses a multi-stage layered progressive method for multi-tenant VM allocation and efficiently handles unprocessed tenants at each stage. This reduces resource fragmentation in cloud data centers, decreases the differences in the QoS among tenants, and improves tenants' overall QoS in cloud data centers. We perform experiments to evaluate the LP-MKP algorithm and demonstrate that it can provide significant gains over other allocation algorithms.

We experimentally evaluate the performance overhead of the virtual environments VMware Player, QEMU, VirtualPC and VirtualBox on a dual-core machine. Firstly, we assess the performance of a Linux guest OS running on a virtual machine by separately benchmarking the CPU, file I/O and the network bandwidth. These values are compared to the performance achieved when applications are run on a Linux OS directly over the physical machine. Secondly, we measure the impact that a virtual machine running a volunteer @home project worker causes on a host OS. Results show that performance attainable on virtual machines depends simultaneously on the virtual machine software and on the application type, with CPU-bound applications much less impacted than IO-bound ones. Additionally, the performance impact on the host OS caused by a virtual machine using all the virtual CPU, ranges from 10% to 35%, depending on the virtual environment.

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.

Energy consumption is significant in cloud data centres even when some virtual machines (VMs) are idle. Taking into account both the energy efficiency and the response performance, we propose a VM allocation strategy with speed switch and VM consolidation in the cloud data centre. When the traffic is lighter, the processing speed of VMs in baseline module will be switched between a low speed and a high speed. When the traffic is much heavier, the VMs in reserved module will be consolidated with the VMs in baseline module, and operates at a high speed. We establish a two-dimensional Markov chain to mathematically estimate the energy-saving degree and the response performance. Numerical experiments with analysis and simulation show that our proposed VM allocation strategy can effectively reduce the energy consumption as well as guarantee the response performance.

We propose two VM migration methods that improve the total sum of program correct by dynamically moving VMs to appropriate edge node. We assume a multi-hierarchy information processing system that the cloud data center and the edge node VM (Virtual Machine) perform processing at the same time, the data center completes all processing, and the VM has the required response time and returns a result that the processing has progressed (called accuracy). The first method is “VM sweeping method” that allocates more CPU time to VMs that perform high accuracy processing. The second methods is “VM sweeping method” that allocates more CPU time to VMs that perform high accuracy processing that improves per unit CPU time (AIR), and “VM number averaging method” that attempts to move VMs so that the assignment status of the CPU time in all edge nodes is equal. The accuracy improvement effect of the proposed VM migration method was evaluated by a simulator. Evaluations changed the AIR from the Early-blooming type where the accuracy greatly increases in the first half to the Late-blooming type where the accuracy greatly increases in the second half, and comparing it with the case where the VM is not moved. It was clarified that a big effect can be expected in the VM sweeping method, and the VM number averaging method in the Earlyblooming type program.

Live migration of virtual machine (VM) is a promising technology that helps physical machines (PMs) adapt to load changes and guarantees Quality of Service (QoS) in cloud data center. Many individual-based VM migration studies ignore the association between VMs, resulting in high communication cost. Some research on multiple VMs migration migrates the VM group as a whole, which is likely to result in ineffective migration and increase the network burden. In this paper, a VM migration algorithm based on group selection (VMMAGS) is proposed, which takes into account the migration cost, communication cost, and VM heat to optimize migration performance. The appropriate VM groups are selected as migration options, and the optimal migration scheme is obtained according to the integration cost of partitions of selected VM groups. Extensive experiments show that our algorithm can effectively reduce the migration cost and communication cost, improve the system reliability compared with other related algorithms.

Virtual machine (VM) management is one of the key problems in cloud data centers. Many efforts have been put on finding an (approximate) optimal placement of VMs at physical machines (PMs) to achieve high resource utility. In this paper, we consider to reduce the startup delay of VMs, which is a key issue for improving user experience and resource utility. The basic idea of our work is to cache VM images at PMs so as to reduce the time of VM image retrieval. By handling VM placement and image placement jointly and de-centrally, we design a joint placement algorithm and an associated layered transport VM selection algorithm, so as to start VMs quickly in the meanwhile save some calculation resources. The key point of our algorithms lies in how to accelerate the startup of VM with limited information in a distributed fat-tree network. The proposed algorithms are evaluated by extensive simulations using a distributed system simulator SimGrid. The results show that our algorithms just lost a little performance in most cases, compared with the centralized algorithm and a classical cache placement strategy.

Nowadays, Cloud Service Providers (CSPs) provide flexible architectures and reliable internet accessories in cloud datacenters, where different types of applications are hosted. In order to achieve higher resource utilization and better quality of service, virtualization is widely used and implemented with virtual machines (VMs). However, most studies focus on VM management for utilization and energy, not yet considering the cost for emerging internet applications. Therefore, to further reduce network cost, we propose utilization-and-network-driven (UND) and only-network-driven (OND) strategies on top of the common minimal-migration (MM) based VM management as UND-MM and OND-MM. According to simulation results, both UND-MM and OND-MM strategies effectively reduce network cost by averagely 7% and 37%, respectively, of the original MM while retaining comparable resource utilization in CPU and memory for normally-loaded and overloaded datacenters. Moreover, the OND-MM, which is simpler, outperforms the UND-MM, and thus becomes more preferable for VM management in cloud datacenters.

[Purpose] Server virtualization technology is the basic form for the cloud data center to serve customers. While the cloud data center reduces the SLA violations in the early stage, the increase in the utilization of resources such as virtual machine servers and network equipment and decrease in energy consumption have become an important issue for reduction of operating costs. Therefore, on the background of the cloud data center network based on SDN technology, this paper studies a cloud data center resource utilization and energy consumption optimization strategy through virtual machine migration and integration. [Method] The technology based on SDN is firstly introduced, the approximate optimal solution FFD algorithm for the multidimensional vector bin packing problem is then presented, the Partheno-Genetic Algorithm (PGA) is then recommended, finally a Hybrid Partheno genetic Algorithm (HPGA) based on the both is established and a mathematic model is created. [Results] By combining the advantages of FFD and PGA, this paper improves the utilization of the resources of the entire cloud data center, reduces energy consumption, and prevents the occurrence of computing “hot spots”. [Conclusion] Experiments in CloudSim prove the effectiveness of the algorithm. Compared with FFD and PGA algorithms, it is able to enhance the efficiency of the Data Center effectively, and obtain the optimal solution more quickly, thereby balancing the physical resources of the server.

Cloud computing providers have to deal with the energy-performance trade-off: minimizing energy consumption, while meeting service level agreement (SLA) requirements. This paper proposes a new heuristic approach for the dynamic consolidation of virtual machines (VMs) in cloud data centers. The fast best-fit decreasing (FBFD) algorithm for intelligent VMs allocating into hosts and dynamic utilization rate (DUR) algorithm for utilization space and VM migration are successfully proposed. We performed simulations using PlanetLab and GWDG data center workloads to compare our approach against the existing models. It has been observed that the FBFD heuristic algorithm produces better results compared to modified BFD algorithm in terms of energy consumption and SLA violation. Additionally, the time complexity of FBFD algorithm is significantly improved from the order of O(\(m\,*\,n\)) to O(\(m\,*\,\log _2{n}\)). Furthermore, leaving some rates of capacity in the physical machines by the proposed DUR algorithm for VMs to be extended reduces the number of migrations which in turn improves the energy consumption and SLA violation. Our heuristic approach is evaluated using CloudSim and the results show that it performs better than the current state-of-the-art approaches.

Cloud computing model along with virtualization enables scalable infrastructures and provides on demand access to computing resources. Virtualization technology facilitates multiple virtual servers to run on a physical server to enhance the utilization of the server. Hypervisor as a virtualization layer provides an infrastructural support to multiple virtual machines above it by virtualizing hardware resources such as CPU, Memory, Disk and NIC. Hypervisors give near native performance by negating virtualization overhead. ESXi is one of the hypervisors which gives near native performance using full virtualization technique. It is encouraging to study how different Operating Systems (OSs) perform in a virtualized environment in the Private Cloud with ESXi hypervisor. Guest Operating Systems do come in para-virtualized and hardware virtualized flavors. It is innovative idea to compare them from performance perspective to give an idea about their performance variations in different tests. This paper conducts different performance tests with low, medium, high workloads and results are gathered using SIGAR API (System Information Gatherer and Reporter) and explains the behavior of each operating system and gives the reasoning of results with the help of different virtualization techniques. In the experiment, CloudStack (open source cloud computing software) is used to create a private cloud, in which management server is installed on Ubuntu 12.04 - 64 bit operating system. ESXi hypervisor is installed on bare metal as a host in the cluster. The different guest operating systems are installed on ESXi hypervisor and their performances have been evaluated in detail by using SIGAR Framework.