Fine-grained multilayer virtualized systems analysis

Nowadays, Cloud Computing acts a major role in every field. These days, more large data centers are in service and many small cloud data centers are enlarging all over the universe. Cloud Computing is a catchword in the domain of HPC and offers on-demand services to the resources on the internet. The VMs (Virtual Machines) specified in the cloud data centres may have different specifications and instable resource usage, which causes imbalanced resource utilization within servers. Thus, it leads to performance degradation. Hence to achieve efficient selection of VM, these challenges must be addressed and solved by using meta-heuristics algorithms. In order to process the data, the VMs are placed on the PMs (Physical Machines). There will be multiple and dynamic request of input in the IaaS(Infrastructure as a Service) framework, hence the system’s responsibility is to create a VMs without knowing the types of tasks. Therefore, the fixed tasks scheduling is not right for this system. The most important research area that needs to be addressed is its performance in scheduling. The best and optimal solution is to find out in the cloud environment. Metaheuristics-based algorithms provide the near-optimal solution. In this paper, we proposed an Improved Particle Swarm Optimization algorithm to reduce the makespan and improve the throughput. We have compared our results with adaptive three-threshold energy-aware (ATEA) algorithm and PSO. The investigational results display the proposed Improved PSO algorithm will schedule and balance the load in the dynamic cloud environment better than the other approaches.

SummaryVirtual clusters (VCs) have exhibited various advantages over traditional cluster computing platforms by virtue of their extensibility, reconfigurability, and maintainability. As such, they have become a major execution environment for cloud‐based cluster applications. However, compared with traditional clusters, their distributed‐memory programming paradigm still remains largely unchanged, which implies that cluster applications cannot be efficiently deployed in VCs, especially when virtual machines (VMs) are running in different physical hosts. Recently, some efforts have been made to improve inter‐VM communication, resulting in many studies on how cluster applications could take advantages of VCs. However, most of them mainly focus on the situation that the VMs are all coresident on the same physical machine where the message passing mechanism is usually optimized away by exploiting the host's shared memory. In this paper, we present a design and implementation of Naplus, a kernel‐based virtual machine approach to the inter‐VM communications that are across different physical hosts. Naplus is based on Nahanni, a mechanism for shared‐memory communication in virtual environments. As such, it not only inherits the major merits of Nahanni with respect to flexible data structures and efficient synchronization but also achieves a shared‐memory paradigm among VMs. With Naplus, we enable the size of shared space to be maximized as large as the sum of each machine's local memory to accommodate cluster applications with large memory footprints. We prototype Naplus in a dual‐host system where an empirical study is conducted to show the effectiveness of the Naplus approach in achieving distributed shared memory for VCs in data centers. Copyright © 2017 John Wiley & Sons, Ltd.

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.

Modern cloud data centers typically exploit management strategies to reduce the overall energy consumption. While most of the solutions focus on the energy consumption due to computational elements, the optimization of network-related aspects of a data center is becoming more and more important, considering also the advent of the Software-Defined Network paradigm. However, an enabling step to implement network-aware Virtual Machine (VM) allocation is the knowledge of data exchange patterns. In this way we can place in well-connected hosts (or on the same physical host) the couples of VMs that exchange a large amount of information. Unfortunately, in Infrastructure as a Service data centers, a detailed knowledge on VMs data exchange is seldom available without the deployment of a specialized (and costly) monitoring infrastructure. In this paper, we propose a technique to infer VMs communication patterns starting from input/output network traffic time series of each VM. We discuss both the theoretical aspect of such technique and the design challenges for its implementation. A case study is used to demonstrate the viability of our idea.

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.

Cloud computing (CC) could be a massive distributed computing driven by business, during which the services and resources are area unit delivered on request to external consumer via the Web. The distributed computing environment comprises of physical servers, virtual machines, data centers, and load balancers which are appended in an efficient way. With the increasing size of a number of physical servers and utilization of cloud services in data centers (DC), the power consumption is a critical and challenging research problem. Minimizing the operational cost and power in a DC becomes essential for cloud service provider (CSP). To resolve this problem, we introduced a novel approach that leads to nominal operational cost and power consumption in DCs. We propose a multi-objective modified differential evolution algorithm for first placement of virtual machine (VM) in the physical hosts and optimize the power consumption during resource allocation using live migration. The experimental results reveal that our proposed method is significantly better against state-of-the-art techniques in terms of limited power consumption and SLA for any given workload.

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.

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).

In virtualized datacenters (vDCs), dynamic consolidation of virtual machines (VMs) is used to achieve both energy-efficiency and load balancing among different physical machines (PMs). Using VM live migrations, we can consolidate VMs on a smaller number of hosts to power down unused PMs and save energy. Most migration schemes are however oblivious to the characteristics of services that run inside VMs, and thus may lead to migrations where VMs competing for the same resource type are packed on the same PM. As a result, VMs may suffer from significant resource contention and noticeable degradation in their performance. Using resource sensitivity values of VMs (i.e., quantitative measures to reflect how much a VM is sensitive to its requested resources such as CPU, Mem, and Disk), we have designed a novel VM consolidation approach to optimize placement of VMs on available PMs. We validated our approach using five well-known applications/benchmarks with various resource demand signatures: varying from pure CPU/Mem/Disk-intensive to mixtures of them. Our extensive numerical evaluation illustrates that, for the same power consumption, our approach improve the performance of cloud services by 9 - 12%, on average, when compared with current sensitivity oblivious approaches.

Virtualization, one of the most actively adopted technologies today in computer systems, is increasingly widening its range of applications. As multiple virtual machines are concurrently executed on a physical machine, they compete for physical resources such as CPU, DRAM, HDD and NIC. In large scale data centers or cluster systems, therefore, it is required to balance each physical machine's load by migrating a virtual machine to another physical machine. Since system performance is significantly hindered during migration, a technique that can reduce the migration overheads is required. Interestingly, we notice that virtual machines' memory contain numerous dispensable data. Therefore, in this paper, we propose an efficient virtual machine migration technique called D3 (Discarding Dispensable Data). The basic idea is simple but effective: finding indispensable page frames, and then, transferring only those page frames. In D3, simple metadata is handed over for the dispensable page frames, instead of actual data transmission, and the discarded page frames are reconstructed appropriately after the migration process. One issue is that checking whether a page frame is dispensable or not based on comparing the entire content yields excessive overhead. To overcome this problem, we devise a novel sampling based dispensable page frame detection technique. D3 have implemented in Xen virtualization software and evaluated on a realistic environment with relevant workloads. We show that D3 can efficiently reduce migration time up to 64.2% and size of memory required for completing a migration up to 94.8%. We also present that D3 is very useful to provide a certain level of QoS and better performance.

Cloud computing data centers contain a large number of physical machines (PMs) and virtual machine (VMs). This number can increase the energy consumption of the data centers especially when the VMs placed inappropriately on the PMs. This paper presents a new VM placement approach with the objective of minimizing the total energy consumption of a data center. VM placement problem is formulated as a combinatorial optimization problem. Since this problem has been proven to be an NP hard problem, Ant Colony Optimization (ACO) algorithm is adopted to solve the formulated problem. Information heuristic of ACO is used differently based on PM energy efficiency. Experimental results show that the proposed approach scales well on large data centers and significantly outperforms selected benchmark (ACOVMP) in terms of energy consumption.

A multi-objective approach for energy-efficient and reliable dynamic VM consolidation in cloud data centers

By employing the virtual machines (VMs) consolidation technique at a virtualized data center, optimal mapping of VMs to physical machines (PMs) can be performed. The type of optimization approach and the policy of detecting the appropriate time to implement the consolidation process are influential in the performance of the consolidation technique. In a majority of researches, the consolidation approach merely focuses on the management of underloaded or overloaded PMs, while a number of VMs could also be in an underload or overload state. Managing an abnormal state of VM results in the postponement of PM getting into an abnormal state as well and affects the implementation time of the consolidation process. For the aim of optimal VM consolidation in this research, a self-adaptive architecture is presented to detect and manage underloaded and overloaded VMs /PMs in reaction to workload changes in the data center. The goal of consolidation process is employing the minimum number of active VMs and PMs, while guaranteeing the quality of service (QoS). Assessment criteria of QoS are two parameters including average number of requests in the PM buffer and average waiting time in the VM. To evaluate these two parameters, a probabilistic model of the data center is proposed by applying the queuing theory. The assessment results of the probabilistic model form a basis for decision-making in the modules of the proposed architecture. Numerical results obtained from the assessment of the probabilistic model via discrete-event simulator under various parameter settings confirm the efficiency of the proposed architecture in achieving the aims of the consolidation process.

In this paper, we propose a cost effective hybrid SDN-Ethernet network to realize a large data center network. The network can support hundreds of thousands of physical machines (PM). More than 64K tenants can share in a data center and each tenant can individually own 4K VLANs. In addition, each tenant user fully owns the right to assign IP addresses and VLAN IDs to their Virtual Machines (VMs). The network can support VM live migration and fast failure recovery. To enhance survivability, dual SDN controllers are used to manage the network through in-band control. We apply MPTCP to provide multiple concurrent sub-flows to enhance throughput of TCP connections between VMs. A novel congestion aware routing is proposed for network load balancing. We have conducted experiments and simulations to evaluate the performance of the proposed network. The experimental results show that the system can resume connections between VMs in a short time after VM migrations. As a network failure occurs, the proposed failure recovery scheme can restore network connection fast even if the failure disrupting both control and data channels. Our experimental results also reveal that the proposed light weight congestion aware MPTCP routing outperforms the existing ECMP-based and VLB-based routing in two well-known data center topologies.

In modern data centres (DCs), energy consumption and resource management are challenging tasks. So in large virtualized DCs, performance of applications is highly dependent on energy-aware DC architecture and smooth network communication between virtual machines (VMs), while minimizing the communication burden to avoid congestion, latency etc. The communication cost of a network can be reduced by minimizing the VMs migration between physical machines (PMs). Therefore customer as well as service providers need cloud computing solutions that not only minimize operational costs, but also total network load. Minimizing the use of energy/network communication overhead with maximum resource utilization in large DCs is a challenging task as computing applications and data are growing so quickly for which increasing larger servers and disks are required to process them fast enough within the required time period. In order to minimize network load and maximize resource utilization in cloud DCs, we simulated our proposed network load-aware scheduling algorithm that ensures minimum VMs migration while delivering the negotiated Quality-of-Service (QoS). Our simulation also shows results with different bit rates.