Infrastructure as a service

Cloud computing is a new computing paradigm that, just as electricity was firstly gener-ated at home and evolved to be supplied from a few utility providers, aims to transform com-puting into an utility. It is being forecasted that more and more users will rent computing as a ...

To save energy, physical machines (servers) in cloud data centers are partitioned in different pools, depending on whether they are kept on at all times and/or whether they have virtual machines instantiated. Partitioning (pooling) of servers affects the power consumption of the data center but also the performance and responsiveness to user requests. In this paper we examine the behavior of pool management scheme in different operating regions which correspond to linear operation, transition to saturation, and saturation, in particular the tradeoff between performance and power consumption. In addition, our results show that the offered load (analogous to the concept often used in networking) does not offer complete characterization of data center operation; instead, the impact of task arrival rate and task service time must be considered separately.

As more businesses shift to cloud services and mobile devices, will the desktop computer still be as important in the future, or will it slowly fade out?
The desktop computer is unlikely to disappear completely. While mobile devices and cloud services have become dominant for daily tasks and convenience, desktops will remain important for specific, high-performance use cases. The future will likely see a hybrid approach where all three technologies coexist, with each serving a distinct purpose.

Never has the need for robust, resilient organizations been so evident as in recent times with more and more well-established, respected organizations becoming unstable or even perishing as a result of the global recession. Added to these challenging economic times we have the demands of managing increasingly complex organizations, which are often highly dependent on sophisticated Information Systems and technology. Perrow’s Normal Accident Theory (1994) points to the fact that “no matter how hard we try there will be serious accidents because of the interactive complexity”[1] of the organizations in which we work. The introduction of BS25999, the British Standard for business continuity management, draws on international good practice in this field and brings together a clear view of what should constitute effective business continuity management for organizations across all sectors. This has provided Business Continuity Managers with a benchmark against which this aspect of their ...

Virtual Private Server (VPS) enables user access to an operating system instance with a fine-grained control of private software and hardware resources. Many various factors can affect VPS performance and they primarily include physical server hardware specifications, installed operating system, virtualization layer, and the underlying network infrastructure. Therefore, it is very important to properly select a VPS host that meets users, applications and services resource and performance requirements. This paper presents a performance evaluation of three popular VPS hosts; namely Digital Ocean, Linode and VULTR. Performance measurement was conducted under the same controlled conditions for all three VPS hosts using a popular benchmark application for Unix operating systems - UnixBench. Three performance evaluation experiments with a focus on examining and studying key performance metrics which include CPU scheduling, memory management, hard disk drive management and Unix operating s...

In this paper we tackle the problem of providing Quality of Service guarantees to virtualized applications, focusing on computing and networking guarantees. We propose a mechanism for providing temporal isolation based on a CPU real time scheduling strategy. This allows not only to have control over the individual virtual machine throughput, but also on the activation latency and response-time by which virtualized software components react to external events. We show experimental results gathered on a real system validating the approach. ⋆ The research leading to these results has received funding from the European Community's Seventh Framework Programme FP7 under grant agreements n. 214777 "IRMOS-Interactive Realtime Multimedia Applications on Service Oriented Infrastructures" and n. 248465 "S(o)OS -Service-oriented Operating Systems.

was intended as an event to prospect the applications supported by the new paradigm and validate the techniques and the mechanisms. A complementary target was to identify the open issues and the challenges to fix them, especially on security, privacy, and inter-and intra-clouds protocols. We take here the opportunity to warmly thank all the members of the CLOUD COMPUTING 2014 Technical Program Committee, as well as all of the reviewers. The creation of such a high quality conference program would not have been possible without their involvement. We also kindly thank all the authors who dedicated much of their time and efforts to contribute to CLOUD COMPUTING 2014. We truly believe that, thanks to all these efforts, the final conference program consisted of top quality contributions. Also, this event could not have been a reality without the support of many individuals, organizations, and sponsors. We are grateful to the members of the CLOUD COMPUTING 2014 organizing committee for th...

The rapid growth of cloud computing, both in terms of the spectrum and volume of cloud workloads, necessitate re-visiting the traditional rack-mountable servers based datacenter design. Next generation datacenters need to offer enhanced support for: (i) fast changing system configuration requirements due to workload constraints, (ii) timely adoption of emerging hardware technologies, and (iii) maximal sharing of systems and subsystems in order to lower costs. Disaggregated datacenters, constructed as a collection of individual resources such as CPU, memory, disks etc., and composed into workload execution units on demand, are an interesting new trend that can address the above challenges. In this paper, we demonstrated the feasibility of composable systems through building a rack scale composable system prototype using PCIe switch. Through empirical approaches, we develop assessment of the opportunities and challenges for leveraging the composable architecture for rack scale cloud datacenters with a focus on big data and NoSQL workloads. In particular, we compare and contrast the programming models that can be used to access the composable resources, and developed the implications for the network and resource provisioning and management for rack scale architecture.

The rapid growth of cloud computing both in terms of the spectrum and volume of cloud workloads brought many challenges to the traditional data center design -fast changing system configuration requirements due to workload constraints, varying innovation cycles of system components, and maximal sharing of systems and subsystems. In this paper, we developed a qualitative assessment of the opportunites and challenges for leveraging disaggregated datacenter architecture to address these challenges. In particular, we compare and contrast the programming models that can be used to access the disaggregated resources, and developed the implications for the network and resource provisioning and management.

Cloud computing is an emerging market place in which computing resources are treated as the utilities and priced for their usage. A huge competition prevails among the cloud Service Providers (SPs) to offer eco-friendly and economically viable cloud services. Heterogeneous Cluster System (HCS) forms the backbone of such computing resources and for eco-friendly services, minimal energy consumption in HCS is a challenging issue. It not only brings down the operational cost but enhance the system reliability and other environmental veneration as well. Commonly, a green Service Level Agreement (SLA) is made between cloud users and SP. Green SLA ensures that a certain percentage of of the schedule is sacrificed for an energy efficient schedule. This work proposes a scheduling framework to execute heterogeneous independent jobs on a HCS. It also introduces a SLA based negotiation scheme for makespan, energy and cost reffered as "Makespan, Energy and Cost Negotiator (MECN)". The proposed model utilizes all possible empty slacks of the schedule running on Processing Elements (PEs) and extended (due to SLA). Using Dynamic Voltage and Frequency Scaling (DVFS) technique, it provides an energy efficient schedule. A resource replacement strategy is also incorporated, in the proposed framework, to generate a cost effective schedule without sacrificing its performance. The performance of the model is analyzed by its simulation which reflects its usefulness and effectiveness.

Due to the difficulty of employing real data centres' infrastructure for assessing the effectiveness of energy-aware algorithms, many researchers resort to use simulation tools. These tools require precise and detailed models for virtualized data centres in order to deliver accurate results. In recent years, many models have been proposed, but most of them either do not consider energy consumption related to virtual machine (VM) migration or do not investigate some of the energy impacting components (e.g. CPU, network, storage). We propose a new model for data centre energy consumption that takes into account the previously omitted components and provides more accurate energy consumption predictions compared to other state-of-the-art solutions. We evaluate our model's accuracy in a comprehensive set of scenarios implemented in the combined GroudSim/DISSECT-CF simulator. Our analysis revealed a significant improvement in accuracy (up to 42.5%) in the modelled energy consumption compared to a similar state-ofthe-art simulator.

The COMETA e-Infrastructure running in Sicily is compliant with the EGEE middleware and specifications offering a great computing power and huge storage capacity. Since its beginning one of the main goals has been to extend the adoption of Grid paradigm from the academic to business world. Several software and hardware extensions have been implemented in order to enhance the infrastructure performances; they include a new low-latency net layer reserved to heavy parallel applications; some modifications to the parallel job submission and execution procedures for a better support of MPI-based applications; new tools for job monitoring and file catalogue interaction; a scheduling policy tailored on the requirements of a complex environment hosting heterogeneous jobs; the GridLM license server able to grant access to commercial software only to authorised users; finally, the Secure Storage Service defends from insider abuse and completes a very high security level environment. The propo...

Intrusion Detection and Prevention Systems (IDPS) are used: to identify possible attacks, collecting information about them and the trying to stop their occurrence and at last reporting them to the system administrator. These systems are used by some organizations to detect the weaknesses in their security policies, documenting existing attacks and threats and preventing an individual from violating security policies. Because of their advantages these systems became an important part of the security infrastructure in nearly every organization. In a Cloud computing environment, attackers can determine the vulnerabilities in the cloud systems and compromise the virtual machines to set out large scale Distributed Denial-of-Service (DDOS) attack. To avert these virtual machines from concession, we propose a multi-phase solution NICE (Network Intrusion Detection and Countermeasure selection in Virtual Network Systems).

The increasing of energy consumptions and attempts to use Management and Control System based on cloud computing technologies for ensuring normal work of the critical energy infrastructure (CEI) are made serious demands for analyzing Infrastructure as a Service (IaaS) Cloud availability level. In this paper, a new perspective unified platform for determining of availability level of the IaaS Cloud pertaining to the information components of the CEI is described. In fact Semi-Markov Modeling Process for assess of overall availability level of the IaaS Cloud with multiple pools of physical and virtual machines based on a novel Availability as a Cloud Service platform in order to improve effectiveness of CEI was presented by authors.

Results of researches in different areas of science and technique confirm that an effective route in order to solve important tasks is the possibility of migrating data resource within the cloud. Therefore Cloud Computing services including Infrastructure as a Service Cloud (IaaS Cloud) should have high availability level. It is serious problem, because even large cloud providers face with sudden failures of IaaS Cloud. It comes no surprise, that scientists consider taxonomy of this system on base of using overall underlying components, such as physical machines (PMs) and virtual machines (VMs). However cloud providers should always remember that hidden failures of PMs are one of the main causes of damage for their cloud assets. In this paper we propose approach on base of using Semi-Markov model in order to determine availability level for the IaaS Cloud with Technical State Control System.

This research analyzes four cloud service models-IaaS, PaaS, SaaS, and DaaS-using criteria-based evaluations to select the most tailored cloud service model for the scenario of automotive industry manufacturing, cloud-connected automotive manufacturing machines, and the second scenario of global automotive distribution network. Automotive manufacturing needs data infrastructure and general applications for integrated data management, analysis, and sharing. Global distribution networks need global, cloud-connected devices and data. IaaS and DaaS present clear advantages in automotive manufacturing machine evaluation. PaaS and SaaS provide advantages in automotive vehicle distribution network evaluation. A clear advantage exists in the distribution network scenario for the custom evaluation model using local mini-clouds for central distribution center-hub geotopology. The technological requirement and usage of the solutions is the same across both automotive research areas of cloudconnected manufacturing solutions and global distribution networks. Original cloud models are adapted to new research areas of global data analysis, sharing, and infrastructure. Adapting cloud service models for models, definition, evaluation, simulation, and usage is an unexplored, innovative field in research. This research is cross-disciplinary, resulting in a specialization within both strong industries of automotive and cloud computing, offering an environment for further research. These results have the potential for preventing accidents, vehicle recalls, fine-tuning error recovery of a car among relatives, in-situ simulations of autonomous cars, improved repair solutions, predictive diagnostics, searching for lower oil consumption, and performance and fuel usage improvement using data from distributive networks. Such innovations would keep costs low, environmental relations favorable, and the aggressive global automotive industry on the edge. This paper introduces cloud computing and highlights its projected capabilities and impacts in various industries, including the manufacturing sector. The paper also reviews the cloud and its architectures, layers, and cloud computing in the manufacturing sector, detailing the concept of cloud manufacturing and what cloud manufacturing means to manufacturers. Cloud computing is the new generation service-based solution approach that specializes in better service provision for customers. Three main dimensions of cloud computing are significant in understanding the architecture design. First, the nature of the service offered includes IaaS, PaaS, and SaaS. Second, the way service provision for customers is accomplished. Third, the way customers enter and leave the service is offered.

To efficiently design deduplication, caching and other management mechanisms for virtual machine (VM) images in Infrastructure as a Service (IaaS) clouds, it is essential to understand the level and pattern of similarity among VM images in real world IaaS environments. This paper empirically analyzes the similarity within and between 525 VM images from a production IaaS cloud. Besides presenting the overall level of content similarity, we have also discovered interesting insights on multiple factors affecting the similarity pattern, including the image creation time and the location in the image's address space. Moreover, we found that similarities between pairs of images exhibit high variance, and an image is very likely to be more similar to a small subset of images than all other images in the repository. Groups of data chunks often appear in the same image. These image and chunk "clusters" can help predict future data accesses, and therefore provide important hints to cache placement, eviction, and prefetching.

Nigeria is a West African country, often referred to as the "Giant of Africa" due to its population of over 177 million and an estimated GDP of over US$587 billion; abundant natural resources including one of the largest natural gas and crude oil reserves. Despite these strong fundamentals, Nigeria has found it difficult to convert her unique advantages into economic and social benefits for the country. This necessitates diversification of the economy which will require huge investments in infrastructure. In China, Nigeria has found a partner and financier. Despite the importance of Chinese finance for African infrastructure, relatively little is known and or ascertain about its socio-economic value and impacts. The aim of this research work is to identify and discuss the socio-economic impacts of Chinese Government Finances in Infrastructural Development Projects in Nigeria. Secondary sources of data were used to review literature. The result shows that Infrastructural Development Projects in the long run have direct influence on socio-economic development, and the majority of economic activities depend one way or the other on road networks, airports, railway transportation systems and power generation projects.

The recent increase in availability of Infrastructure-as-a-Service (IaaS) computing clouds provides a new way for researchers to run complex scientific applications. However, using cloud resources for a large number of research jobs requires significant effort and expertise. Furthermore, running jobs on many different clouds presents even more difficulty. In order to make it easy for researchers to deploy scientific applications across many cloud resources, we have developed a virtual machine resource manager (Cloud Scheduler) for distributed compute clouds. In response to a user's job submission to a batch system, the Cloud Scheduler manages the distribution and deployment of user-customized virtual machines across multiple clouds. We describe the motivation for and implementation of a distributed cloud using the Cloud Scheduler that is spread across both commercial and dedicated private sites, and present some early results of scientific data analysis using the system.

We show that distributed Infrastructure-as-a-Service (IaaS) compute clouds can be effectively used for the analysis of high energy physics data. We have designed a distributed cloud system that works with any application using large input data sets requiring a high throughput computing environment. The system uses IaaS-enabled science and commercial clusters in Canada and the United States. We describe the process in which a user prepares an analysis virtual machine (VM) and submits batch jobs to a central scheduler. The system boots the user-specific VM on one of the IaaS clouds, runs the jobs and returns the output to the user. The user application accesses a central database for calibration data during the execution of the application. Similarly, the data is located in a central location and streamed by the running application. The system can easily run one hundred simultaneous jobs in an efficient manner and should scale to many hundreds and possibly thousands of user jobs.

Cloud computing represents a new way to deliver and use services on a shared IT infrastructure. Traditionally, IT hardware and software were acquired and provisioned on business premises. Software applications were built, possibly integrating off-the-shelf components, deployed and run on these privately owned resources. With service-oriented computing, applications are offered by service providers to clients, who can simply invoke them through the network. The offer specifies both the functionality and the Quality of Service (QoS). Providers are responsible for deploying and running services on their own resources. Cloud computing moves one step further. Computing facilities can also be delivered on demand in the form of services over a network. In this paper we take the perspective of a Software as a Service (SaaS) provider whose goal is to maximize the revenues from end users who access services on a pay-per-use basis. In turn, the SaaS provider exploits the cloud, which provides an Infrastructure as a Service (IaaS), where the service provider dynamically allocates hardware physical resources. This paper presents a distributed algorithm for run-time management of SaaS cloud systems that jointly addresses the capacity allocation and admission control of multiple classes of applications providing an heuristic solution which closely approximates the global optimal solution.

Cloud computing offers a powerful abstraction that provides a scalable, virtualized infrastructure as a service where the complexity of fine-grained resource management is hidden from the end-user. Running data analytics applications in the cloud on extremely large data sets is ...

Infrastructure cloud systems offer basic functionalities only for managing complex virtual infrastructures. These functionalities demand low-level understanding of applications and their infrastructural needs. Recent research has identified several techniques aimed at enabling the semi-automated management and using applications that span across multiple virtual machines. Even with these efforts however, a truly flexible and end-user oriented approach is missing. This paper presents the One Click Cloud Orchestrator that not only allows higher level of automated infrastructure management than it was possible before, but it also allows end-users to focus on their computational problems instead of the complex cloud infrastructures needed for their execution. To accomplish these goals the paper reveals the novel building blocks of our new orchestrator from the components closely related to infrastructure cloud to the ways virtual infrastructures are modeled. Finally, we show our initial evaluation and study on how the orchestrator fulfills the high level requirements of end-users. The research leading to these results has received funding from the European Community 's Seventh Framework Programme FP7/2007-2013 under grant agreement no. 608886 (CloudSME)

Infrastructure cloud systems offer basic functionalities only for managing complex virtual infrastructures. These functionalities demand low-level understanding of applications and their infrastructural needs. Recent research has identified several techniques aimed at enabling the semi-automated management and using applications that span across multiple virtual machines. Even with these efforts however, a truly flexible and end-user oriented approach is missing. This paper presents the One Click Cloud Orchestrator that not only allows higher level of automated infrastructure management than it was possible before, but it also allows end-users to focus on their computational problems instead of the complex cloud infrastructures needed for their execution. To accomplish these goals the paper reveals the novel building blocks of our new orchestrator from the components closely related to infrastructure cloud to the ways virtual infrastructures are modeled. Finally, we show our initial evaluation and study on how the orchestrator fulfills the high level requirements of end-users. The research leading to these results has received funding from the European Community 's Seventh Framework Programme FP7/2007-2013 under grant agreement no. 608886 (CloudSME)

1 formation models and resource class hierarchies simplify resource descriptions from multiple providers, pathfinding and topology embedding algorithms rely on query abstractions as building blocks. The paper describes how the semantic resource description models enable ExoGENI to autonomously instantiate on-demand virtual topologies of virtual machines provisioned from cloud providers and are linked by on-demand virtual connections acquired from multiple autonomous network providers to serve a variety of applications ranging from distributed system experiments to high-performance computing.

The center of Cloud processing incorporates virtualization of equipment assets, for example, stockpiling, system and memory gave through virtual machines. The live relocation of these VMs is acquainted with get different advantages which chiefly incorporate high accessibility, equipment support, blame takeover and outstanding task at hand adjusting. Other than different offices of the virtual machine relocation, it is powerless to extreme safety dangers amid movement process because of which the business is reluctant to acknowledge it. The exploration done till date is on the execution of movement process; while the safety perspectives in relocation are not completely investigated. We have completed a broad review to research the vulnerabilities, dangers and conceivable assaults on the live virtual machine migration. Besides, we have recognized security necessities for safe virtual machine relocation and exhibited a nitty gritty examination of current arrangements based on these sec...

In this paper, we describe an analytical modeling approach for performability analysis of cloud provided services. We use infrastructure-as-a-service as an example of a cloud based service, where service availability and provisioning response delays are key service level agreements (SLAs) to users while cost reductions are important to service providers. A novelty of our approach is in reducing the complexity of the analysis by dividing the overall model into multiple interacting sub-models and then obtaining the overall solution by iteration over submodel solutions. This results in a high fidelity model that is scalable. We summarize our modeling approach, showcase our ongoing work through initial results and outline future avenues of research using such an approach.

Choosing a proper divergence measure is a common task for a statistician and probabilist. The Kullback-Leibler divergence is well known among the information divergence. We introduce an upper bound on the Kullback-Leibler divergence. We show that this upper bound is better than the χ 2 bound. Based on the proposed bound, we find a two sided bound for Shannon's entropy. A simulation study shows that the new bound works to choose a proper model among the rival models. It is illustrated on the logistic regression and some non-nested models.

We present a new systems runtime, EbbRT, for cloud hosted applications. EbbRT takes a different approach to the role operating systems play in cloud computing. It supports stitching application functionality across nodes running commodity OSs and nodes running specialized application specific software that only execute what is necessary to accelerate core functions of the application. In doing so, it allows tradeoffs between efficiency, developer productivity, and exploitation of elasticity and scale. EbbRT, as a software model, is a framework for constructing applications as collections of standard application software and Elastic Building Blocks (Ebbs). Elastic Building Blocks are components that encapsulate runtime software objects and are implemented to exploit the raw access, scale and elasticity of IaaS resources to accelerate critical application functionality. This paper presents the EbbRT architecture, our prototype and experimental evaluation of the prototype under three d...

IaaS clouds are major enablers of data-intensive cloud applications because they provide necessary computing capacity for managing Big Data environments. In a typical IaaS cloud, virtual machine (VM) instances deployed on physical machines (PM) are provided to the users for their computing needs. Recently, IaaS cloud providers are realizing that merely providing the basic functionalities for Big Data processing is not sufficient to survive intense business competitions. Rather, the performance of the cloud provided service is an equally important factor when a

We demonstrate the application of optical combs to implement tunable programmable microwave photonic filters. We design well-known multi-tap microwave photonic filters; however, the utilization of an optical comb with a dispersive medium enables scaling of these filters to a large number of taps. We use optical line-by-line pulse shaping to program tap weights, which allows us to shape the filter's bandpass. Our scheme is simple and easily implementable, which provides filters with arbitrary tap weights. As an example, we implement filters with Gaussian apodized tap weights, which achieve more than 35-dB sidelobe suppression. Our experiments provide usable bandwidth, free of sampling spurs, over a Nyquist zone of 5 GHz, equal to half of our 10-GHz comb repetition frequency. Furthermore, we introduce a simple new technique, based on a programmable optical delay line, to uniformly tune the passband center frequency across the free spectral range (FSR) of the filter, ideally without changing the bandpass shape. We demonstrate this scheme by tuning the filter over a full FSR, equal to 10.4 GHz in our experiments.

Cloud computing is internet based technology that provides online resources on pay per use basis. This computing technology is dependent on virtualization. By registering with cloud providers customers can use services (soft and hard) on pay per use basis. Customers are getting services in the form of virtual machines, they interact with these virtual machines to do their jobs. In recent survey it is found that security is the top most concern associated with cloud services. Customers are always afraid of adopting clouds services as their data are stored out of their network boundary and administration. Researchers in cloud are trying to develop security models to protect cloud services from intrusions including internal and external. In this paper we have proposed an intrusion security model that uses the optimum cloud resources and increases the resource and service availability.

Cloud computing offers reduced capital expenditure, operational risks, complexity and maintenance, and increased scalability while providing services at different abstraction levels, namely Software-as-a-Service (SaaS), Platform-as-a-Service (PaaS), and Infrastructure-as-a-Service (IaaS). A new approach called cloud networking adds networking functionalities to cloud computing and enables dynamic and flexible placement of virtual resources crossing provider borders. This allows various kinds of optimization, e.g., reducing latency or network load. However, this approach introduces new security challenges. This paper presents a security architecture that enables a user of cloud networking to define security requirements and enforce them in the cloud networking infrastructure.

We present a simulation study of several prefetching policies to improve the I O performance of external merging using parallel I O. In particular we consider traditional sequential prefetch, forecast-based greedy prefetching, and oblivious prefetching. In conjunction with the prefetching policies we e v aluate the bene t of two di erent data placement strategies: run-level striping and block-random placement, in the presence of data skew. We show that the I O performance is greatly improved by using forecasting techniques. This method outperforms the other policies in achieving higher disk parallelism, and scales well with increased numbers of disks and increasing data skew. Additionally, the performance of block-random data placement i s s h o wn to be uniformly good, independent of the data skew, while the performance of run-level striped placement degrades with increasing skew.

Nigeria is a West African country, often referred to as the "Giant of Africa" due to its population of over 177 million and an estimated GDP of over US$587 billion; abundant natural resources including one of the largest natural gas and crude oil reserves. Despite these strong fundamentals, Nigeria has found it difficult to convert her unique advantages into economic and social benefits for the country. This necessitates diversification of the economy which will require huge investments in infrastructure. In China, Nigeria has found a partner and financier. Despite the importance of Chinese finance for African infrastructure, relatively little is known and or ascertain about its socioeconomic value and impacts. The aim of this research work is to identify and discuss the socioeconomic impacts of Chinese Government Finances in Infrastructural Development Projects in Nigeria. Secondary sources of data were used to review literature. The result shows that Infrastructural Development Projects in the long run have direct influence on socioeconomic development, and the majority of economic activities depend one way or the other on road networks, airports, railway transportation systems and power generation projects.

Grids organize resource sharing, a fundamental requirement of large scientific collaborations. Seamless integration of grids into everyday use requires responsiveness, which can be provided by elastic Clouds, in the Infrastructure as a Service (IaaS) paradigm. This paper proposes a model-free resource provisioning strategy supporting both requirements. Provisioning is modeled as a continuous action-state space, multi-objective reinforcement learning (RL) problem, under realistic hypotheses; simple utility functions capture the high level goals of users, administrators, and shareholders. The model-free approach falls under the general program of autonomic computing, where the incremental learning of the value function associated with the RL model provides the so-called feedback loop. The RL model includes an approximation of the value function through an Echo State Network. Experimental validation on a real dataset from the EGEE grid shows that introducing a moderate level of elasticity is critical to ensure a high level of user satisfaction.

Simulation tools play a critical role in various domains of cloud computing. They offer researchers and infrastructure designers a virtual environment to work in, eliminating the expenses associated with physical infrastructure, so those platforms could offer determinant support to analyze the performance of cloud systems and the related data centers. For those reasons, many cloud simulators have been developed. They are currently used by academics and practitioners, but only a few of these provide useful features in order to perform effective simulations of cloud systems for energy-related issues. The goal of this work is to help academics and professionals interested in energy-aware simulations of cloud systems to select the most suitable simulation platform for their purposes. The first part of this paper reviews the five simulators that hold the most promising features regarding energy issues. Subsequently, an evaluation metric has been introduced, based on five qualitative indexes: Last update, Popularity, Availability, Granularity, and Performance profile. The evaluation aims to provide researchers with suggestions for the best simulator choice.

Cloud computing is a normal evolution of distributed computing combined with Service-oriented architecture, leveraging most of the GRID features and Virtualization merits. The technology foundations for cloud computing led to a new approach of reusing what was achieved in GRID computing with support from virtualization. The Third International Conference on Cloud Computing, GRIDs, and Virtualization (CLOUD COMPUTING 2012), held between July 22 and 27, 2012, in Nice, France, intended to prospect the applications supported by the new paradigm and validate the techniques and the mechanisms. A complementary target was to identify the open issues and the challenges to be fixed, especially on security, privacy, and inter- and intra-clouds protocols. We welcomed technical papers presenting research and practical results, position papers addressing the pros and cons of specific proposals, such as those being discussed in the standard fora or in industry consortia, survey papers addressing t...

Helix Nebula-the Science Cloud Initiative-is a public-private partnership between Europe's leading scientific research organisations and European IT cloud providers. CERN contributed to this initiative by providing a flagship use case: the workloads from the ATLAS experiment. Aiming to gain experience in managing and monitoring large-scale deployments, as well as in benchmarking the cloud resources, a sizable Monte Carlo production was performed using the Helix Nebula platform. This contribution describes the Helix Nebula initiative and summarizes the experience and the lessons learned from deploying ATLAS experiment application within large cloud setups involving several commercial providers.

Cloud computing infrastructures enable companies to cut costs by outsourcing computations on-demand. However, clients of cloud computing services currently have no means of verifying the confidentiality and integrity of their data and computation. To address this problem we propose the design of a trusted cloud computing platform (TCCP). TCCP enables Infrastructure as a Service (IaaS) providers such as Amazon EC2 to provide a closed box execution environment that guarantees confidential execution of guest virtual machines. Moreover, it allows users to attest to the IaaS provider and determine whether or not the service is secure before they launch their virtual machines.

Cloud Computing is a promising paradigm in deployment of software and there survive. Computer researcher says a magnificent sentence, cloud computing is next generation operating system. Technologies such as cluster, grid, and now, cloud computing, have all aimed at allowing entrance to large amounts of computing power in a fully virtualized, paravirtulized manner. The Energy efficiency of information and communication technology becomes more and more vital due to elevate of energy costs and the world wide desire to decrease CO2 emissions [13],[14]. In this paper we argues energy-efficient job scheduling and allocation scheme that minimize number of hosts, so amount of energy conserve.

Modern block-oriented distributed storage systems like Hadoop distributed file system have proliferated in this era of big data and cloud computing. These systems feature block-level replication in which their files are partitioned into equalsized blocks and multiple copies for each block are then arbitrarily distributed across nodes for fault tolerance and data availability. However, many storage volumes are just wasted only for keeping block copies whose data may not be accessed frequently in the strategy. Therefore, distributed storage systems begin to adopt erasure codes. However, classical parity encoding scheme are hard to be directly applied to the distributed storage systems since block copies are arbitrarily placed across nodes in the systems. We present a novel technique, called DynaEC, to address the issues in modern block-oriented distributed storage systems. DynaEC provides a unique parity encoding algorithm that encodes data blocks arbitrarily distributed across machines to parities and then places the parities guaranteeing fault tolerance. Parity encoding in DynaEC is performed without any change of the original block placement policy in Hadoop distributed file system. This makes DynaEC work seamlessly with Hadoop distributed file system. Finally, during the encoding procedure each data node encodes each own data blocks, not requiring any information about other blocks located in other data nodes. As such, the encoding procedure in DynaEC is fully performed in parallel without

Hadoop Distributed File System (HDFS) is the core component of Apache Hadoop project. In HDFS, the computation is carried out in the nodes where relevant data is stored. Hadoop also implemented a parallel computational paradigm named as Map-Reduce. In this paper, we have measured the performance of read and write operations in HDFS by considering small and large files. For performance evaluation, we have used a Hadoop cluster with five nodes. The results indicate that HDFS performs well for the files with the size greater than the default block size and performs poorly for the files with the size less than the default block size.

Hadoop Distributed File System (HDFS) is the core component of Apache Hadoop project. In HDFS, the computation is carried out in the nodes where relevant data is stored. Hadoop also implemented a parallel computational paradigm named as Map-Reduce. In this paper, we have measured the performance of read and write operations in HDFS by considering small and large files. For performance evaluation, we have used a Hadoop cluster with five nodes. The results indicate that HDFS performs well for the files with the size greater than the default block size and performs poorly for the files with the size less than the default block size.

Cloud computing promises ondemand scalability and flexibility for the enterprise environment and open source products have a large presence in this area. We have devised an elaborate set of criteria for evaluation and comparison of open source IaaS cloud computing solutions which can be used for ranking and choosing between the available products. We explain each criteria group, and give both the rationale and the estimated impact of the criteria. We also briefly introduce a number of most common open source cloud computing IaaS solutions to be evaluated using the explained criteria.

Several works have shown that the time to boot one virtual machine (VM) can last up to a few minutes in high consolidated cloud scenarios. This time is critical as VM boot duration defines how an application can react w.r.t. demands’ fluctuations (horizontal elasticity). To limit as much as possible the time to boot a VM, we design the YOLO mechanism (You Only Load Once). YOLO optimizes the number of I/O operations generated during a VM boot process by relying on the boot image abstraction, a subset of the VM image (VMI) that contains data blocks necessary to complete the boot operation. Whenever a VM is booted, YOLO intercepts all read accesses and serves them directly from the boot image, which has been locally stored on fast access storage devices (e.g., memory, SSD, etc.). Creating boot images for 900+ VMIs from Google Cloud shows that only 40 GB is needed to store all the mandatory data. Experiments show that YOLO can speed up VM boot duration 2-13 times under different resourc...