Virtualized Data Centers Research Articles

Software-Defined Networking for Scalable Cloud-based Services to Improve System Performance of Hadoop-based Big Data Applications

The rapid growth of Cloud Computing has brought with it major new challenges in the automated manageability, dynamic network reconfiguration, provisioning, scalability and flexibility of virtual networks. OpenFlow-enabled Software-Defined Networking (SDN) alleviates these key challenges through the abstraction of lower level functionality that removes the complexities of the underlying hardware by separating the data and control planes. SDN has an efficient, dynamic, automated network management, higher availability and application provisioning through programmable interfaces which are very critical for flexible and scalable cloud-based services. In this study, the author explores broadly useful open technologies and methodologies for applying an OpenFlow-enabled SDN to scalable cloud-based services and a variety of diverse applications. The approach in this paper introduces new research challenges in the design and implementation of advanced techniques for bringing an SDN-enabled components and big data applications into a cloud environment in a dynamic setting. Some of these challenges become pressing concerns to cloud providers when managing virtual networks and data centers, while others complicate the development and deployment of cloud-hosted applications from the perspective of developers and end users. However, the growing demand for manageable, scalable and flexible clouds necessitates that effective solutions to these challenges be found. Hence, through real-world research validation use cases, this paper aims at exploring useful mechanisms for the role and potential of an OpenFlow-enabled SDN and its direct benefit for scalable cloud-based services. Finally, it demonstrates the impact of an OpenFlow-enabled SDN that fully embraces the opportunities and challenges of cloud infrastructures to improve the system performance of Hadoop-based big data applications by utilizing the network control capabilities of an OpenFlow to solve network congestion.

Culturally Afforded Tensions in the Second Life Metaverse

This study investigated the culturally contingent tensions afforded by the implementation of Second Life in transatlantic communications among 13 college-level students at a Southwestern academic institution in the United States and their instructor and an assistant professor and his graduate student at a Greek-speaking academic institution. The transatlantic transactions unfolded in IBM's virtual Green Data Center, where students and instructors engaged in critical discussions on their local community, IBM, the European Union, and the United States' sustainability practices. By analyzing students and instructors' virtual exchanges, chat medium, and reflective comments, three categories of culturally enacted tensions were identified. These contradictions pertained to (a) emerging intercultural communication, (b) assigned collaborative activities, and (c) the use of the Second Life Viewer as a communication tool. The study demonstrated that contradictions can be contingent on institutional, broader cultural and historical constructs, Internet-mediated tools, and the different linguistic and cultural values and expectations related to relationship building and interactional dynamics. The study highlighted the importance of acknowledging such structural tensions as affordances for teaching and learning. Further, the study emphasized the need to take into account the institutional, pedagogical, instructional, and broader cultural realisms that impose constraints on teaching practices and participation in social virtualities.

Lightness: A Function-Virtualizable Software Defined Data Center Network With All-Optical Circuit/Packet Switching

Modern high-performance data centers are responsible for delivering a huge variety of cloud applications to the end-users, which are increasingly pushing the limits of the currently deployed computing and network infrastructure. All-optical dynamic data center network (DCN) architectures are strong candidates to overcome those adversities, especially when they are combined with an intelligent software defined control plane. In this paper, we report the first harmonious integration of an optical flexible hardware framework operated by an agile software and virtualization platform. The LIGHTNESS deeply programmable all-optical circuit and packet switched data plane is able to perform unicast/multicast switch-over on-demand, while the powerful software defined networking (SDN) control plane enables the virtualization of computing and network resources creating a virtual data center and virtual network functions (VNF) on top of the data plane. We experimentally demonstrate realistic intra DCN with deterministic latencies for both unicast and multicast, showcasing monitoring, and database migration scenarios each of which is enabled by an associated network function virtualization element. Results demonstrate a fully functional complete unification of an advanced optical data plane with an SDN control plane, promising more efficient management of the next-generation data center compute and network resources.

Workload Consolidation using VM Selection and Placement Techniques in Cloud Computing

Cloud computing provides a consumer pay-per-use computing model over the Internet using numerous data centers across the globe. Power consumption by the huge data centers in Cloud environment has attracted the attention of research community. Efficient usage of energy in Cloud can be addressed in many facets. Virtual Machine (VM) consolidation is one of the techniques to save or reduce energy in virtualized data centers. VM Migration in Cloud also provides us an opportunity for reducing energy consumption. In this research, we intend to study various VM placements & selection policies and VM migration algorithms for underloaded and overloaded hosts to reduce energy consumption and SLA violation. We propose a novel method using combination of two methods, Least Increase Power (LIP) consumption with Host Sort and Minimum Correlation Coefficient (MCC) for consolidation of VM placement, placing a migratable VM on a host based on utilization thresholds. The results show performance of each combination of algorithms varies with the changing value of the parameters brings better in terms of energy consumption, VM migration time and SLA violation. The reader may plunge the appropriate method for energy consumption.

The virtual atomic and molecular data centre (VAMDC) consortium**http://www.vamdc.eu.

The Virtual Atomic and Molecular Data Centre (VAMDC) Consortium is a worldwide consortium which federates atomic and molecular databases through an e-science infrastructure and an organisation to support this activity. About 90% of the inter-connected databases handle data that are used for the interpretation of astronomical spectra and for modelling in many fields of astrophysics. Recently the VAMDC Consortium has connected databases from the radiation damage and the plasma communities, as well as promoting the publication of data from Indian institutes. This paper describes how the VAMDC Consortium is organised for the optimal distribution of atomic and molecular data for scientific research. It is noted that the VAMDC Consortium strongly advocates that authors of research papers using data cite the original experimental and theoretical papers as well as the relevant databases.

Poris: A Scheduler for Parallel Soft Real-Time Applications in Virtualized Environments

With the prevalence of cloud computing and virtualization, more and more cloud services including parallel soft real-time applications (PSRT applications) are running in virtualized data centers. However, current hypervisors do not provide adequate support for them because of soft real-time constraints and synchronization problems, which result in frequent deadline misses and serious performance degradation. CPU schedulers in underlying hypervisors are central to these issues. In this paper, we identify and analyze CPU scheduling problems in hypervisors. Then, we design and implement a parallel soft real-time scheduler according to the analysis, named Poris, based on Xen. It addresses both soft real-time constraints and synchronization problems simultaneously. In our proposed method, priority promotion and dynamic time slice mechanisms are introduced to determine when to schedule virtual CPUs (VCPUs) according to the characteristics of soft real-time applications. Besides, considering that PSRT applications may run in a virtual machine (VM) or multiple VMs, we present parallel scheduling, group scheduling and communication-driven group scheduling to accelerate synchronizations of these applications and make sure that tasks are finished before their deadlines under different scenarios. Our evaluation shows Poris can significantly improve the performance of PSRT applications no matter how they run in a VM or multiple VMs. For example, compared to the Credit scheduler, Poris decreases the response time of web search benchmark by up to 91.6 percent.

Thermal Prediction Models for Virtualized Data center Servers by using Thermal-profiles

The energy dissipated as heat for each utilization level of a data center server is empirically measured and stored as the thermal-profile. These thermal-profiles are used to predict the outlet temperatures of the related servers for current and future utilization. The predicted outlet temperature is an important parameter for energy efficient thermal-aware workload scheduling and workload migration in green data centers. This paper presents three models for outlet temperature prediction on virtualized data center servers based on thermal-profile. The best case scenario managed to predict the outlet temperature with a negligible error of 0.3 degree Celsius.

Optimal VDC Service Provisioning in Optically Interconnected Disaggregated Data Centers

Virtual data center (VDC) is a key service in modern data center (DC) infrastructures. However, the rigid architecture of traditional servers inside DCs may lead to blocking situations when deploying VDC instances. To overcome this problem, the disaggregated DC paradigm is introduced. In this letter, we present an integer linear programming (ILP) formulation to optimally allocate VDC requests on top of an optically interconnected disaggregated DC infrastructure, aiming to quantify the benefits that such an architecture can bring when compared with traditional server-centric DCs. Moreover, a lightweight simulated annealing-based heuristic is provided for the scenarios where the ILP scalability is challenged. The obtained numerical results reveal the substantial benefits yielded by the resource disaggregation paradigm.

Study of Smart Transportation Data Center Virtualization Based on VMware vSphere and Parallel Continuous Query Algorithm over Massive Data Streams

There are various shortages of the traditional data center in cost, resource utilization, power consumption and operation. This paper studies the development trend of the virtual data center and its technical advantages, proposes a scheme of virtual system of smart transportation data center based on VMware vSphere. At the same time, oriented on requirements of traffic data stream management in city smart transportation system under the environment of cloud computing, aimed at massive, multi-source, real-time, dynamic uncertain data stream sent back from all kinds of cross regional intensive control perception device, this paper analyses the characteristics and correlation of actual city traffic operation and traffic data flow, researches evolution mechanism of uncertain data; construct traffic data flow model based on ontology, core metadata and theory of constraints. And on this basis, this paper uses virtualization and large data sets of parallel processing, considers load balancing and adaptive mechanism, combines fuzzy theory and dynamic multi object, multi constrained decision theory, to seek the efficient query algorithm for dynamic, complex and continuous transportation data stream.

A Cost Effective Virtual Cluster with Hadoop Framework for Big Data Analytics

Big data processing is currently becoming increasingly important research field in computer technology professionals due to the continuous growth of the amount of data generated by various fields. However, the processing of large-scale research data requires cluster technology infrastructure which causes huge investments for educational institutions. Hadoop is an open-source framework that allows for distributed storage and processing of very large data sets on computer clusters built from commodity hardware. This technology is being widely used for the analysis of large datasets. This paper focuses on proposes a low cost scalable hadoop virtual cluster platform and the performance of hadoop irtual cluster. We first describe the design and implementation of a virtual datacenter using hadoop framework. Then we perform a set of experiments to investigate the performance of virtual datacenter with standard datasets. For experiment and evaluation, the Cloudera's distribution of Apache Hadoop (CDH) is installed and configured on the DIU Cloud and Big Data Lab as a prototype implementation of a virtual data center and tested with 40 workstations. In this paper, we use TPC BENCHMARK ™ DS” by-Transaction Processing Performance Council (TPC) benchmarks for evaluation virtual data center performance. The contributions of this paper, is to design model and implement a cost effective elastic virtual data center with hadoop framework and resource utilizations for educational institutions to provide high performance for distributed and parallel processing; as well as, Identify the bottlenecks of this systems e.g. bandwidth of network connectivity with all nodes in the lab.

Joint study on VMs deployment, assignment and migration in geographically distributed data centers

Enterprises build private clouds to provide IT resources for geographically distributed subsidiaries or product divisions. Public cloud providers like Amazon lease their platforms to enterprise users, thus, enterprises can also rent a number of virtual machines (VMs) from their data centers in the service provider networks. Unfortunately, the network cannot always guarantee stable connectivity for their clients to access the VMs or low-latency transfer among data centers. Usually, both latency and bandwidth are in unstable network environment. Being affected by background traffics, the network status can be volatile. To reduce the latency uncertainty of client accesses, enterprises should consider the network status when they deploy data centers or rent virtual data centers from cloud providers. In this paper, we first develop a data center deployment and assignment scheme for an enterprise to meet its users' requirements under uncertain network status. To accommodate to the changes of the network status and users' demands, a VMs migration-based redeployment scheme is adopted. These two schemes work in a joint way, and lay out a framework to help enterprises make better use of private or public clouds.

Parallel virtual machine migration in WDM optical data center networks

Using virtualization in data centers (DCs), people can run multiple applications independently at a single location, and also can move the virtual machine (VM) to a suitable place. Since a DC carries plentiful applications, its failure would result in service paralysis. To guarantee the service continuity, the relative VMs should migrate from a failed DC to the target backup one. The wavelength division multiplexing (WDM) technology is a promising solution of providing a high-rate and stable inter-DC VM migration trajectory. Thus, to mitigate the service paralysis caused by DC failures, in this paper, we propose an effective parallel inter-DC VM migration framework with the consideration of VM grading, load balancing and elastic optical bandwidth allocation. Compared with previous end-to-end VM migration approach, our method shrinks migration duration exponentially with the higher migration profit and the lower blocking rate. Finally, the simulation results testify these superiorities, and the improvement ratio arrives up to 43%.

MOL-D: A Collisional Database and Web Service within the Virtual Atomic and Molecular Data Center

MOL-D database is a collection of cross-sections and rate coefficients for specific collisional processes and a web service within the Serbian Virtual Observatory (SerVO) and the Virtual Atomic and Molecular Data Center (VAMDC). This database contains photo-dissociation cross-sections for the individual ro-vibrational states of the diatomic molecular ions and rate coefficients for the atom-Rydberg atom chemi-ionization and inverse electron-ion-atom chemi-recombination processes. At the moment it contains data for photodissociation cross-sections of hydrogen H2+ and helium He2+ molecular ions and the corresponding averaged thermal photodissociation cross-sections. The ro-vibrational energy states and the corresponding dipole matrix elements are provided as well. Hydrogen and helium molecular ion data are important for calculation of solar and stellar atmosphere models and for radiative transport, as well as for kinetics of other astrophysical and laboratory plasma (i.e. early Universe).

Stark Widths of Spectral Lines of Neutral Neon

In order to complete Stark broadening data for Ne I spectral lines which are needed for analysis of stellar atmospheres, collisional widths and shifts (the so-called Stark broadening parameters) of 29 isolated spectral lines of neutral neon have been determined within the impact semiclassical perturbation method. Calculations have been performed for the broadening by collisions with electrons, protons and ionized helium for astrophysical applications, and for collisions with ionized neon and argon for laboratory plasma diagnostics. The shifts have been compared with existing experimental values. The obtained data will be included in the STARK-B database, which is a part of the Virtual Atomic and Molecular Data Center – VAMDC.

On the Stark Broadening of Lu III Spectral Lines

The electron-impact widths for 27 Lu III spectral lines have been calculated by using the modified semiempirical method. Calculations have been also performed with the published relativistic Hartree-Fock oscillator strengths and additionally, with the approximate formula of Cowley. With the obtained results, the influence of Stark broadening on Lu III lines was investigated in the spectra of A-type stars. The obtained data will be included in the STARK-B database, which is part of the Virtual Atomic and Molecular Data Center – VAMDC.

Cooperative Virtual Data Center: Sharing Data and Resources among Multiple Computing Entities

Existing data centers are each individually owned and operated by a single entity. This situation creates an excessive financial burden upon each entity through the need to overprovision for hardware. While this state of affairs enables a secure and an efficient maintenance scheme for the data center, the financial drawbacks are perhaps excessive. To address the concern, we propose the idea of a cooperative virtual data center among multiple entities that is founded on the principal of fair resource sharing amongst the entities.

Sponge: an oversubscription strategy supporting performance interference management in cloud

Resource oversubscription optimizes the utilization of the computing resources. Many well-known virtual machine monitors (VMMs), such as Xen and KVM, adopt this approach to help maximize the yield of the cloud datacenters. That is, with proper resource oversubscription strategies, more virtual machines (VMs) can be supported by limited resources. However, performance interference among VMs hosting in the same physical machines (PMs) exists in cloud environment, and probably aggravated by resource oversubscription strategies, which aims to put more VMs into the same PM. In this paper, we present a resource oversubscription strategy called Sponge targeting cloud platforms. Sponge mitigates the issue of performance interference among the oversubscribed co-hosting VMs. Sponge also provides a VM association strategy for each PM to handle with its besteffort We performed our evaluation on a virtual datacenter simulated by Xen. Our evaluation results show that Sponge improves the resources utilization and manages to make each VM meet its performance requirement even hosting with other VMs in the same PM.

Virtual network function embedding in real cloud environments

Future network architectures will be completely reshaped by the emerging Network Function Virtualization (NFV) paradigm, and telco operators will likely deploy flexible infrastructures based on the cloud, offering programmable connectivity services and computing/storage facilities in the form of Virtual Data Centers (VDCs). This paper introduces and discusses some challenging technical issues associated with this promising approach, with a special focus on the exploitation of industry-relevant cloud platforms, such as OpenStack, and on network-aware optimal placement problem of entire VDCs, taking into account multiple virtual (Virtual Machines – VMs, virtual networks, …) and physical (host capacities, network topology and capacity, …) resources and constraints. The proposed placement, computed for real production OpenStack deployments, not only satisfies the predicted communication for all VDCs deployed atop the same physical data center, but also accounts for the (real) computing overhead due to intense communication load of VMs co-located on the same physical host, such as in the case of Virtual Network Functions (VNF) embedding, typically neglected by similar existing works.

A new technique for efficient live migration of multiple virtual machines

Datacenter virtualization technologies have attracted a lot of attention to enable various cloud computing services and to facilitate virtual machine (VM) migration. VM migration can help service providers to achieve the goals of saving energy, enhancing resource efficiency and quality of service (QoS). In order to ensure the QoS, the migration time and the downtime of VM should be considered while implementing the VM migration. Most researches focus on the issue of single VM migration by using the post-copy migration strategy or pre-copy migration strategy. However, there is few research that focuses on the problem of live migration for multiple VMs. Therefore, in this paper, we first propose an improved serial migration strategy and introduce the post-copy migration scheme into it. We then propose the m mixed migration strategy that is based on the improved serial migration strategy and the parallel migration strategy. Furthermore, we develop queuing models (i.e., the M/M/C/C and the M/M/C queuing models) to quantify performance metrics, such as the blocking ratio and average waiting time of each migration request. We evaluate the performance of the proposed migration strategy by conducting mathematical analysis, the numerical results show that our proposed strategy outperforms the existing approach.

Resource Harvesting in Cognitive Wireless Computing Networks With Mobile Clouds and Virtualized Distributed Data Centers: Performance Limits

We consider a virtualized data center (VDC) consisting of a set of servers hosting a number of mobile terminals forming a mobile cloud and study the problem of optimal resource allocation in the presence of time varying workloads and uncertain channels. The channel uncertainty may be due to fading and/or uncertain link availability and reliability in cognitive wireless networks. The servers are processing certain applications delegated to them by the terminals, for either energy saving or due to the lack of necessary software at the terminal to process the applications. The control problem is to dynamically adjust resources according to channel and workload fluctuations in order to maximize the long-term average throughput and to minimize the energy cost of the overall system while maintaining network stability. We develop a unified VDC model for both cognitive and conventional wireless networks, carry out a unified stability analysis, and characterize the joint stability region for the unified VDC model. We also propose a new dynamic policy that supports every point in the network stability region, outperforms previously proposed network stabilizing policies without using the information of arrival statistics, and mitigates the mutual impact of primary and secondary service providers on each other.