China
Africa
Explore chapters and articles related to this topic
2
Published in Carlos Alberto Vélez Quintero, Optimization of Urban Wastewater Systems using Model Based Design and Control, 2020
Amazon EC2 offers four different types of instances named as small, medium, large and extra large; depending on the characteristics of the machine. The first experiment was the evaluation of the performance of those instances associated with the UWwS optimization using a sequential NSGAII algorithm. The benchmark model was set up in such a way that in each instance the objective function was evaluated 25 times in a single core. For comparison, a computer from the local network was also evaluated. Table 6.3 shows the characteristics of the instances available in Amazon EC2 and the desktop computer called “local”. The results of the sequential optimization experiment are presented in Figure 6.10 and Table 6.4.
Cloud computing for big data
Published in Jun Deng, Lei Xing, Big Data in Radiation Oncology, 2019
Amazon EC2 provides a virtual computing environment (IaaS) that enables users to run VMs on Amazon’s Cloud. Users can either create new Amazon machine images (AMI) containing an operating system (Windows or Linux), applications, and libraries or select from a collection of globally available AMIs. AMIs are stored using the S3, Amazon’s distributed Cloud storage, as objects grouped in buckets. Each bucket can be stored in one of several geographical regions. The regions can be specified by users considering latency, cost, and regulatory requirements. Tasks are billed by Amazon EC2 according to the run time (CPU hour used) and the level of performance (CPU speed, memory, disk storage) of the machine selected by the user. Jobs handled by S3, on the other hand, are charged according to the amount of data stored and transferred (GB per month). CloudWatch is another service provided by Amazon that delivers performance metrics such as CPU utilization pattern, network transfer volume, or disk operations out of the data generated by other services like Amazon EC2. Finally, Amazon Virtual Private Cloud (VPC) securely connects a user’s private infrastructure to Amazon’s Cloud through a virtual private network (VPN). This enables owners of an already setup infrastructure to enhance their network options by connecting to Amazon Web Services (AWS) while being protected by Amazon’s security services, firewalls, and intrusion detection systems.
Web Services Delivered from the Cloud
Published in John W. Rittinghouse, James F. Ransome, Cloud Computing, 2017
John W. Rittinghouse, James F. Ransome
Amazon EC2 presents a true virtual computing environment, allowing clients to use a web-based interface to obtain and manage services needed to launch one or more instances of a variety of operating systems (OSs). Clients can load the OS environments with their customized applications. They can manage their network’s access permissions and run as many or as few systems as needed. In order to use Amazon EC2, clients first need to create an Amazon Machine Image (AMI). This image contains the applications, libraries, data, and associated configuration settings used in the virtual computing environment. Amazon EC2 offers the use of preconfigured images built with templates to get up and running immediately. Once users have defined and configured their AMI, they use the Amazon EC2 tools provided for storing the AMI by uploading the AMI into Amazon S3. Amazon S3 is a repository that provides safe, reliable, and fast access to a client AMI. Before clients can use the AMI, they must use the Amazon EC2 web service to configure security and network access.
Pricing the cloud: a QoS-based auction approach
Published in Enterprise Information Systems, 2020
Yang Lu, Xianrong Zheng, Ling Li, Li D. Xu
Cloud computing, an evolutionary technology in the ICT industry, integrates several technologies including distributed computing, grid computing, and virtualization technology for enterprise applications (Mell and Grance 2011; Li et al. 2012; Zhang et al. 2016). Cloud services include Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS), etc. (Armbrust et al. 2010; Xu et al. 2017, 2019). Currently, leading IT companies provide cloud services for their customers, such as Amazon AWS (Amazon Web Services), Google’s GAE1 (Google App Engine), Microsoft’s Azure2, and IBM’s Cloud. In terms of pricing mechanisms, Amazon Elastic Compute Cloud (Amazon EC2) offers seven instance purchasing options3: On-demand Instances, Reserved Instances, Scheduled Instances, Spot Instances, Dedicated Hosts, Dedicated Instances, and Capacity Reservations. However, Amazon EC2 and other cloud services have not yet provided a pricing mechanism that incorporates QoS directly. In other words, the relationship between its price and QoS has not been explored.
Pareto-Based Adaptive Resources Selection Model in Hybrid Cloud Environment
Published in IETE Journal of Research, 2021
Ketaki Bhalchandra Naik, G. Meera Gandhi, S. H. Patil
This article is influenced from our previous paper [1]. The new information originates from the following aspects. (1) In the earlier paper, NSGA-II (non-dominated sorting GA-II) and GSA (Gravitational Search Algorithm) relied on the selection of virtual machine from private cloud whereas in this article, Adaptive Multiobjective Resource Selection Model (AMORSM) uses A-NSGA-2 with GSA algorithm for selecting the resources from the hybrid cloud environment. (2) Here we have used the adaptive acceleration coefficient to maintain the diversity among the frontiers. (3) In addition to this, AMORSM model is also capable of handling unpredictable workloads and managing the execution of the tasks within a deadline. (4) Also, we have used the elitist NSGA-II with niching to maintain the Pareto optimal schedule until the completion of iteration. (5) Similarly, this article has significantly enhanced the experimental evaluation. We have compared the performance trade-off with the state-of-the-art multiobjective algorithms. Many cloud applications experience dynamic workloads at various time scales during its execution. The cloud platforms have the ability to allocate, compute and storage resources promptly to handle workload fluctuations. Most of the time companies have their own on-premises private data center to carry out their work. However, unexpected workload spikes can sometimes increase the need of resources as compared to locally available resources. In addition to this, if a user submits an application with deadline constraint to already overloaded private cloud, then the application generates demand for extra resources. In such type of scenario, company uses the hybrid cloud model [2] by supplementing the external cloud resources for the execution of an application rather than incurring capital expenditure on deployment cost and operation cost for additional resources within premises. Thus the hybrid cloud model enhances the dynamic provision of resources and minimizes the hardware cost. The exponential need of virtualization has made companies to avail the unlimited resources from external cloud providers such as IBM smart cloud [3], Amazon EC2 [4] by creating Virtual Machines (VM) that allow the organization to upsurge its scope as needed. In virtue of this, the model needs to know the real-time status of the infrastructure of the private as well as public clouds to register them with information repositories to process the resource discovery, monitoring and selection operations. In this context, we develop an AMORSM to select resources to respond quickly and efficiently to infrequent peak workloads within the user-specified deadlines. AMORSM uses a multiobjective evolutionary algorithm called A-NSGA2 with GSA for the Pareto optimal selection of resources from the hybrid cloud.