Overall 5G-MoNArch architecture and implications for resource elasticity

With diverse and rich offerings from cloud computing providers in the open cloud market, scientists have great opportunities to design and conduct complex applications by utilizing and combining computational resources, software components and data sources in elastic manners. While existing techniques focus mainly on resource elasticity in single cloud infrastructure, scientists expect to design their applications being elastic in multiple dimensions to ensure that they applications can operate on multiple clouds with minimum software engineering effort. In this paper we will focus on providing techniques for scientists to compose elastic high performance applications by utilizing traditional software components, user-provided components and cloud services. We characterize elastic compositions via their resource, quality, cost, available time and usage right elasticity, thus enabling scientists to evaluate and decide how to develop, deploy and control the compositions to match their elastic needs. To illustrate our approach, we will present several real-world application compositions for multi-cloud environments.

Distributed data stream processing and edge computing: A survey on resource elasticity and future directions

In present days cloud computing is one of the greatest platform which provides storage of data in very lower cost and available for all time over the internet.But it has more critical issue like security, load management and fault tolerance. In this paper we are discussing Load Balancing approach. Resource scheduling management design on Cloud computing is an important problem. Scheduling model, cost, quality of service, time, and conditions of the request for access to services are factors to be focused. A good task scheduler should adapt its scheduling strategy to the changing environment and load balancing Cloud task scheduling policy. Cloud Computing is high utility software having the ability to change the IT software industry and making the software even more attractive. It has also changed the way IT companies used to buy and design hardware. The elasticity of resources without paying a premium for large scale is unprecedented in the history of IT industry. The increase in web traffic and different services are increasing day by day making load balancing a big research topic. Cloud computing is a new technology which uses virtual machine instead of physical machine to host, store and network the different components. Load balancers are used for assigning load to different virtual machines in such a way that none of the nodes gets loaded heavily or lightly. The load balancing needs to be done properly because failure in any one of the node can lead to unavailability of data.

Cloud technology has revolutionized the landscape of computing in the last decade. Benefits such as reduced costs, rapid application deployment, and elastic resources, have enticed many organizations to utilize cloud resources or host much of their data in the cloud. Recent studies have shown that over 70 percent of the world's businesses now operate at least some of their operations in the cloud, with many more expected to join in the coming years. In the past, security concerns deterred many organizations from using cloud computing services. In this paper a review of three critical cloud security threats is discussed, with the purpose of determining if the new widespread adoption of cloud computing is due to advances in security. The review revealed that although new security technologies have been introduced, many issues concerning the three threats are still present today. Because of this, continued widespread adoption of cloud computing will likely increase compromises until innovative techniques are created to address cloud computing threats.

Adoption of any new technology in an organization is a crucial decision as it can have its impact at technical, economical, and organizational level. One of such decisions is related to adoption of Cloud-based services in an organization. Cloud Computing provides elastic resources as per the demand and provides the facility to pay as per the use. Thus, it is changing the way IT infrastructure is used today with huge benefit of cost savings. However, if the solution adopted by an organization is not fulfilling the requirements, it can have tremendous negative consequences at technical, economical, and organizational level. Therefore, the decision to adopt Cloud-based services should be based on a methodology that supports a wide array of criteria for evaluating the available alternatives. Also, as these criteria or factors can be mutually interdependent and conflicting, a trade-offs-based methodology is needed to make such decisions. This paper, therefore, discusses the design, implementation, and evaluation of the prototype developed for automating the theoretical methodology of Trade-offs based Methodology for Adoption of Cloud-based Services (TrAdeCIS) developed in (Garg and Stiller, 2014). This system is based on Multi-attribute Decision Algorithms (MADA), which selects the best alternative, based on the priorities of criteria of decision maker. In addition the applicability of this methodology to the adoption of cloud-based services in an organization is validated with several use-cases towards the end of the paper. Furthermore, the extendibility of this system to other domains is being evaluated with respect to Train Operating Companies, who wish to find out the best alternative of providing Internet connectivity and voice calls on-board trains.

Cloud computing is the promising key technology to build future architecture of massive IT systems and one of key benefits of cloud computing is to provide its customers with elastic resources according to the fluctuation of request workloads. In this paper, we propose adaptive resource management policy to handle requests of deadline-bound application with elastic cloud. Adaptive resource management architecture has been proposed, and we divide resource management into two parts, resource provision and job scheduling. We design analytical provision model for adaptive provision based on queuing theory, by introducing a key metric named average interval time. Three job scheduling policies are raised to dequeue appropriate jobs to execute, First-Come-First-Service (FCFS), Shortest Job First (SJF) and Nearest Deadline First (NDF), for different preference toward execution order. Simulation evaluation has been set up with realistic grid workload, and results show that our provisioning model gives elastic resource provisioning for dynamic workload and FCFS achieves better performance compared with other scheduling policies.

In large-scale machine-to-machine sensor networks, the applications such as urban air pollution monitoring require information management over widely distributed sensors under restricted power, processing, storage, and communication resources. The continual increases in size, data generating rates, and connectivity of sensor networks present significant scale and complexity challenges. Traditional schemes of information management are no longer applicable in such a scenario. Hence, an elastic resource allocation strategy is introduced which is a novel management technique based on elastic computing. With the discussion of the challenges of implementing real-time and high-performance information management in an elastic manner, an air pollution monitoring system, called EIMAP, was designed with a four-layer hierarchical structure. The core technique of EIMAP is the elastic resource provision scheduler, which models the constraint satisfaction problem by minimizing the use of resources for collecting information for a defined quality threshold. Simulation results show that the EIMAP system has high performance in resource provision and scalability. The experiment of pollution cloud dispersion tracking presents a case study of the system implementation.

As open source private cloud platform, OpenStack provides basic resource service which has features of stability reliability and security. Function components of OpenStack platform were deployed and installed. Integrated basic resource pool was built. Based on open API of platform, applying JAVA native interface, elastic resource provisioning system was implemented which adops software architecture of B/S. The system has functions of establishment distribution and real-time adjustment of dynamic resource. In order to manage elastic resource of plateform expediently the system provides personalization module and favorable user interface.

The level of participation in research by GPs is low internationally. Previous reports of the reasons why practitioners decline opportunities for research participation have tended to recount the barriers that they describe as if they are objective accounts. By theoretical sampling of practitioners who had declined to participate in a research trial, we sought to interpret the functional significance and interrelationship of the barriers that they reported. Twenty-three GPs who had declined to participate in a trial of training to manage medically unexplained symptoms were interviewed and their accounts analysed interpretatively. The practitioners described general practice and research as alien fields. Research lacked intrinsic, clinical or professional value and was linked to evidence-based medicine which they rejected as incompatible with person-centered care. Every doctor described a lack of time for research, but time was an elastic resource that payment could release from the reservoir of their 'own time'. The findings should inform the design and interpretation of future quantitative surveys to identify how common the attitudes that we report are. Doctors with the attitudes of those whom we interviewed will not be drawn into research by measures predicated on the assumption that it is intrinsically, clinically or professionally valuable. If they cannot be convinced of its utility, value could be conferred by payment for participation.

Load balancing is the core of virtual resource management and scheduling in cloud computing. For network applications, the cost of user would be greatly saved if load balancer could dynamically adjust cluster resources in accordance with the current applied load. The current load balancing products of cloud, such as Amazon's ELB, can be used to manage virtual machines in the cloud. The main drawbacks are still only supporting template-based deployment of new virtual machines, not supporting the trend prediction, failing to gain resources dynamically, and not sufficiently providing the elastic management of resources. Since the virtual machine for load balancing management in cloud computing can be dynamically applied and released, an algorithm of prediction-based elastic load balancing resource management (TeraScaler ELB) is presented to overcome the drawbacks. Experiments have shown that the required number of virtual machines change in compliance with the change of network load, thus TeraScaler ELB is able to dynamically adjust the processing capacity of back-end server cluster with the applied load. Besides it could make full use of the 'use on demand' feature of cloud computing, TeraScaler ELB leads to a better application of prediction based load balancing in cloud computing. It concludes that compared with the traditional elastic resource management algorithm, TeraScaler ELB is more reasonable for providing scalability and high availability.

In a distributed stream processing system, elastic resource provisioning/scheduling is the main factor that affects system performance and limits system applications. However, in the data stream computing platform, resource allocation is often suboptimal due to the large fluctuations of the data stream rate, which creates a performance bottleneck for the cluster. In this paper, we propose a data stream prediction strategy (Dp-Stream) for elastic computing system to mitigate the resource allocation issue. First, we establish a back propagation (BP) neural network prediction model based on genetic simulated annealing algorithm to predict the trend of the data stream rate in the next time window of the cluster; second, according to the time latency, the estimation model adjusts the resources allocated to the critical operations of the critical path in the Directed Acyclic Graph (DAG) and finally, the resource communication cost is optimized. We evaluate the prediction accuracy and system latency of the proposed scheduling strategy in Storm. The experimental results prove the feasibility and effectiveness of the proposed strategy.KeywordsData stream predictionResource schedulingStream computingBack propagation neural networkStorm

Resource elasticity and server consolidation have long been among two of cloud computing’s most relevant management tools. Yet, exemplified with a scientific application use case, this paper highlights how judicious scheduling of tasks can help maximize resource utilization and improve performance and costs for both users and cloud providers. Developing an efficient cloud service for DNA sequence comparisons is adopted as a motivating use case. Using the bioinformatics tool MASA that finds an optimal pair-wise sequence alignment, we propose a model for co-scheduling multiple alignments on a single cloud instance. The resulting, practically optimal, non-preemptive schedule can effectively double the throughput of MASA-based sequence alignment workflows.

Today's complex cloud applications are composed of multiple components executed in multi-cloud environments. For such applications, the possibility to manage and control their cost, quality, and resource elasticity is of paramount importance. However, given that the cost of different services offered by cloud providers can vary a lot with their quality/performance, elasticity controllers must consider not only complex, multi-dimensional preferences and provisioning capabilities from stakeholders but also various runtime information regarding cloud applications and their execution environments. In this chapter, the authors present the elasticity control approach of the EU CELAR Project, which deals with multi-dimensional elasticity requirements and ensures multi-level elasticity control for fulfilling user requirements. They show the elasticity control mechanisms of the CELAR project, from application description to multi-level elasticity control. The authors highlight the usefulness of CELAR's mechanisms for users, who can use an intuitive, user-friendly interface to describe and then to follow their application elasticity behavior controlled by CELAR.

A resource elasticity framework for QoS-aware execution of cloud applications

Deep Learning-based (DL) applications are becoming increasingly popular and advancing at an unprecedented pace. While many research works are being undertaken to enhance Deep Neural Networks (DNN)-the centerpiece of DL applications-practical deployment challenges of these applications in the Cloud and Edge systems, and their impact on the usability of the applications have not been sufficiently investigated. In particular, the impact of deploying different virtualization platforms, offered by the Cloud and Edge, on the usability of DL applications (in terms of the End-to-End (E2E) inference time) has remained an open question. Importantly, resource elasticity (by means of scale-up), CPU pinning, and processor type (CPU vs GPU) configurations have shown to be influential on the virtualization overhead. Accordingly, the goal of this research is to study the impact of these potentially decisive deployment options on the E2E performance, thus, usability of the DL applications. To that end, we measure the impact of four popular execution platforms (namely, bare-metal, virtual machine (VM), container, and container in VM) on the E2E inference time of four types of DL applications, upon changing processor configuration (scale-up, CPU pinning) and processor types. This study reveals a set of interesting and sometimes counter-intuitive findings that can be used as best practices by Cloud solution architects to efficiently deploy DL applications in various systems. The notable finding is that the solution architects must be aware of the DL application characteristics, particularly, their pre- and post-processing requirements, to be able to optimally choose and configure an execution platform, determine the use of GPU, and decide the efficient scale-up range.