Cloud-based Data Centers Research Articles

CPU Provisioning Algorithms for Service Differentiation in Cloud-Based Environments

This work focuses on the design, analysis and evaluation of Dynamic Weighted Round Robin (DWRR) algorithms that can guarantee CPU service shares in clusters of servers. Our motivation comes from the need to provision multiple server CPUs in cloud-based data center environments. Using stochastic control theory we show that a class of DWRR policies provide the service differentiation objectives, without requiring any knowledge about the arrival and the service process statistics. The member policies provide the data center administrator with trade-off options, so that the communication and computation overhead of the policy can be adjusted. We further evaluate the proposed policies via simulations, using both synthetic and real traces obtained from a medium scale mobile computing application.

H-PC: a cloud computing tool for supervising hypertensive patients

Hypertension or high blood pressure is a condition on the rise. Not only does it affect the elderly but it is also increasingly spreading to younger sectors of the population. Treating it involves exhaustive monitoring of patients. Current health services can be improved to perform this task more effectively. A tool adapted to the particular requirements of hypertension can greatly facilitate monitoring and diagnosis. This paper presents the computer application Hypertension Patient Control (H-PC), which allows patients with hypertension to send their readings through mobile phone Short Message Service (SMS) or e-mail to a cloud computing datacenter. Through a graphic interface, clinicians can keep track of their patients, thus facilitating monitoring. Cloud-based datacenters provide a series of advantages in terms of scalability, maintainability, and massive data processing. However, the ability to guarantee Quality of Service (QoS) is crucial for the commercial success of cloud platforms. A novel and efficient cloud-based platform managing H-PC with QoS is also proposed in this paper.

Reliable Design for Stochastic Multicast Virtual Network in Data Centres

ABSTRACTNetwork virtualization enables multiple heterogeneous virtual networks (VNs) to coexist on a shared infrastructure. Mapping a VN onto a substrate network is a key issue in building a VN, and has been addressed by intensive research and algorithms. However, current research on designing VNs and VN mapping is suitable both for unicast service-oriented VNs with static or dynamic resource demand and for mapping multicast VNs (MVNs) with deterministic network resource requirement. In this paper, we investigate the online survivable mapping for stochastic MVN (SM-SMVN) requests in cloud-based data centres, in which the bandwidth demand of the MVN request follows a normal distribution, i.e. .In this work, we first model the SM-SMVN problem by using mixed integer linear programming (MILP) with the objective of minimizing mapping cost while considering load balancing. Then, we devise a heuristic algorithm SM-SMVN for tackling this NP-hard problem efficiently. The numerical results obtained from extensive simulation experiments demonstrate the effectiveness and superiority of the proposed approach compared with the traditional over-provisioning solutions.

Efficient Provisioning of Hybrid Virtual Network with Stochastic Resource Demands

ABSTRACTNetwork virtualization is a key technology to support multiple heterogeneous virtual networks (VNs) on a shared substrate infrastructure. Most current research on designing VNs and VN mapping is either efficient for unicast service-oriented VNs with static or dynamic resource demand, or suitable for mapping multicast VNs (MVNs) with deterministic network resource requirement. In this paper, we investigate the online mapping for stochastic hybrid VN (M-SHVN) requests in cloud-based data centres, where the requests consist of both multicast and unicast VN requests. We assume that the bandwidth demand of the hybrid virtual network (HVN) request follows a normal distribution, i.e. . We devise a heuristic algorithm M-SHVN for tackling this NP-hard problem efficiently. The numerical results obtained from our extensive simulation experiments demonstrate the effectiveness and superiority of the proposed approach compared with the traditional “over-provisioning” solutions.

Cloudy-Based Optimization of the Human Resource Management System

With the development of cloud-based data centers, cloud-based system has been widely used in many kinds of enterprises. Human resource management is an important issue in enterprise management. In this paper, we proposed a cloud based system for human resource management. Under the help of our design, the performance of the human resource management system can be improved apparently. In order to illustrate its superiority,we use some experiment to evaluate the proposed system. The experimental results indicated that the cloudy-based system would consume less memory and cpu resources while it has faster connection speed.

Survivable provisioning for multicast service oriented virtual network requests in cloud-based data centers

Network virtualization technology plays an important role in cloud-based data centers that serve as an effective method for provisioning a flexible and highly adaptable shared substrate network to satisfy the demands of various applications. Although there has been some work on efficient mapping of unicast service oriented virtual networks there has been very limited work on addressing the problem of efficient mapping of multicast oriented virtual networks. Furthermore, how to guarantee survivability of multicast service oriented virtual network request has not been studied.In this work, we investigate the survivable multicast service oriented virtual network mapping (SMVNM) problem and propose an efficient algorithm for solving this problem. We first formulate the SMVNM problem with the objective of minimizing mapping cost by using mixed integer linear programming. We then design an efficient algorithm to solve this problem since it is NP-hard. We validate and evaluate our framework and algorithms by conducting extensive simulations on different realistic networks under various scenarios, and by comparing with existing approaches. The simulation results show that our approach outperforms existing solutions.

An energy-efficient enhanced virtual resource provision middleware in clouds

In the past few years, cloud computing has emerged as a promising paradigm for delivering IT-infrastructures, platforms, applications and services. The consolidation of this new paradigm in both commercial business and academic research requires that the underlying resources should be economically managed. In this paper, we present a resource provision framework, which is aiming at improving the energy-efficiency of cloud-based data centres. The proposed framework is implemented as a lightweight middleware, which consists of three novel services that enable cloud providers to up/down scale their virtual resource pool in an efficient manner. Extensive experiments are conducted to evaluate the effectiveness and the performance of the proposed middleware in real-world cloud platform. The results indicate that it can significantly improving the energy-efficiency in virtualised clouds without significantly degradation on application performance.

Who Is Your Neighbor: Net I/O Performance Interference in Virtualized Clouds

User-perceived performance continues to be the most important QoS indicator in cloud-based data centers today. Effective allocation of virtual machines (VMs) to handle both CPU intensive and I/O intensive workloads is a crucial performance management capability in virtualized clouds. Although a fair amount of researches have dedicated to measuring and scheduling jobs among VMs, there still lacks of in-depth understanding of performance factors that impact the efficiency and effectiveness of resource multiplexing and scheduling among VMs. In this paper, we present the experimental research on performance interference in parallel processing of CPU-intensive and network-intensive workloads on Xen virtual machine monitor (VMM). Based on our study, we conclude with five key findings which are critical for effective performance management and tuning in virtualized clouds. First, colocating network-intensive workloads in isolated VMs incurs high overheads of switches and events in Dom0 and VMM. Second, colocating CPU-intensive workloads in isolated VMs incurs high CPU contention due to fast I/O processing in I/O channel. Third, running CPU-intensive and network-intensive workloads in conjunction incurs the least resource contention, delivering higher aggregate performance. Fourth, performance of network-intensive workload is insensitive to CPU assignment among VMs, whereas adaptive CPU assignment among VMs is critical to CPU-intensive workload. The more CPUs pinned on Dom0 the worse performance is achieved by CPU-intensive workload. Last, due to fast I/O processing in I/O channel, limitation on grant table is a potential bottleneck in Xen. We argue that identifying the factors that impact the total demand of exchanged memory pages is important to the in-depth understanding of interference costs in Dom0 and VMM.

Performance Analysis of Network I/O Workloads in Virtualized Data Centers

Server consolidation and application consolidation through virtualization are key performance optimizations in cloud-based service delivery industry. In this paper, we argue that it is important for both cloud consumers and cloud providers to understand the various factors that may have significant impact on the performance of applications running in a virtualized cloud. This paper presents an extensive performance study of network I/O workloads in a virtualized cloud environment. We first show that current implementation of virtual machine monitor (VMM) does not provide sufficient performance isolation to guarantee the effectiveness of resource sharing across multiple virtual machine instances (VMs) running on a single physical host machine, especially when applications running on neighboring VMs are competing for computing and communication resources. Then we study a set of representative workloads in cloud-based data centers, which compete for either CPU or network I/O resources, and present the detailed analysis on different factors that can impact the throughput performance and resource sharing effectiveness. For example, we analyze the cost and the benefit of running idle VM instances on a physical host where some applications are hosted concurrently. We also present an in-depth discussion on the performance impact of colocating applications that compete for either CPU or network I/O resources. Finally, we analyze the impact of different CPU resource scheduling strategies and different workload rates on the performance of applications running on different VMs hosted by the same physical machine.

Optimal provisioning for virtual network request in cloud-based data centers

Today applications and services are migrating to a cloud-computing-based paradigm in which the users access the applications and services hosted in data centers, by using thin-clients on the user terminal device. These applications/services are typically hosted and run on virtual machines in interconnected data centers. Different applications from the same user may need to access and change shared data or information. Thus, we may abstract the applications from same user as a virtual network (VN). For better performance and efficiency, it is critical that the VN request be accommodated with optimal provisioning under the current resource state of data centers. In this paper, for addressing the issue of how to design an optimal provisioning scheme for the VN request such that the total revenue of is maximized, we first develop a framework for the optimal provisioning of VN request by using mixed integer programming. Since the optimal provisioning problem is NP-hard, we also propose a genetic algorithm---based heuristic algorithm for addressing the problem of optimal provisioning for VN with unsplittable flow and optimal provisioning for VN with splittable flow problems. We demonstrate the effectiveness of our approach in improving the total revenue by conducting extensive simulations on different networks.

Reinventing virtual appliances

Today, data centers are being built with more entities on a larger scale in order to satisfy the rapid growth of services and the demand for efficient service delivery and management. These data centers are facing scalability, manageability, and consumability challenges that require breakthrough technology for design and management. This paper extends previous work that focused on using virtual appliances for service management, by describing the following areas: 1) rapid provisioning, which simplifies and accelerates the deployment and activation of virtual appliances by checking parameter dependencies and coordinating execution sequences; 2) selective presentation, which provides flexible interfaces to present meaningful information through customized metrics; 3) simplified operation, which enables different operation granularities by coordinating the operation sequences across virtual appliances; and 4) automated consolidation, which enables automated service performance management by automatically monitoring and optimizing resource allocations. In addition, we describe the design and prototype of an extendable management framework, virtualization integrator (VSI), which enables an appliance builder to easily fulfill the key features of virtual appliances and coordinate management function across components located in multiple virtual appliances. By reinventing virtual appliances with built-in management capability, data center management is radically simplified and automated. The results, improved server utilization and simplified system manipulation, demonstrate the potential for virtual appliances to be fundamental building blocks for large-scale data centers and potentially for cloud-based new enterprise data centers.