Cloud Computing Solutions. Группа авторов
Чтение книги онлайн.
Читать онлайн книгу Cloud Computing Solutions - Группа авторов страница 25
Information is stored in different locations which are regulated by numerous associations: Cloud services are provided by various CSPs, which maintain hardware components, such as processors, memory and storage, in individualized parts. Therefore, the user information is stored in numerous locales which are controlled by a few associations. Thus, the purported “issue of numerous ways” could emerge based on the capacity of specialists and/or people to process and manage the administration by the accumulation, utilization, distribution and revealing of their personal information by others. Boundless universal information utilization, information sharing and pools of information storage among the associations would become an issue in the perplexing association of cloud services. It is very hard to verify who will assume the liability if anything undesirable happens to stored information. All the logging information must be checked and should keep information acquisition, information transmission or information change.
Accountability: This is another part of protection and security. Sufficient data is needed with respect to how information is overseen by the CSPs so they can keep track of information acquisition, information transmission or information change. That is the reason why the representatives of associations authorize the standards and directions used to maintain bookkeeping records.
2.6 Major Cloud Service Providers
This section discusses some popular CSPs, including Amazon, Microsoft, Google, Sun and Yahoo, which epitomize utilization of cloud computing. These CSPs are being used by individual customers having the extensive ventures discussed below.
2.6.1 IaaS CSPs
Amazon Elastic Compute Cloud (EC2): This is a business web service which enables clients to lease personal computers on which to run and convey their personal computer applications. EC2 permits versatile arrangement of office applications by providing a web services interface and an organization platform over which a client can make virtual machines (VM), for example, server cases, where the client can put and run the product of their choice.
Amazon announced a limited public beta test of EC2 on August 25, 2006. Access to EC2 was allowed on a first-come, first-served basis. For the most part, EC2 turned out to be accessible on October 23, 2008 with the help of Microsoft Windows Server.
Amazon.com facilitates customer access with EC2 cloud as one of a few web services advertised under the sweeping title Amazon Web Services (AWS).1 It started its service as IaaS clouds with a variety of services, including S3 (for storage), CloudFront (content delivery), SimpleDB (for structured data storage), SQS (for reliable messaging), CloudFront Streaming (for video streaming), RDS (for relational database), and Elastic MapReduce (for information processing).
Small Instance: The small instance (default) likens to “a framework with 1.7 GB of memory, 1 EC2 Compute Unit (1 virtual center with 1 EC2 Compute Unit), 160 GB of storage space, 32-bit framework.”
Large Instance: The large instance depicts “a framework with 7.5 GB of memory, 4 EC2 Compute Units (2 virtual cores with 2 EC2 Compute Units each), 850 GB of storage space, 64-bit framework.”
Extra Large Instance: The extra-large occurrence offers “what might as well be called a framework with 15 GB of memory, 8 EC2 Compute Units (4 virtual cores with 2 EC2 Compute Units each), 1690 GB of storage space, 64-bit framework.”
Amazon charges customers in two primary ways:
Hourly cost/VM.
Information exchange cost.
The hourly VM rate is arrived at in view of the limit and enhancements of the VM. Amazon publicizes the evaluation plan as “pay for resources you consume.”
Clients can begin and stop VMs to manage costs without much trouble, with Amazon estimating with hourly granularity. Few are ready to keep each virtual machine running close to the limit and pay just for CPU time really used.
FlexiScale: This is a UK-based CSP that offers services which are equivalent to Amazon Web Services. The virtual servers of this cloud provider offer features such as persistent storage by default, dedicated VLAN, fixed IP address, runtime adjustment of CPU, and a wider range of server sizes. FlexiScale cloud services are also priced by the hour.
In the current scenario in the UK, FlexiScale cloud is providing services such as Web service (SOAP), access to virtual server through SSH (Linux), web-based user interfaces, Remote Desktop (Windows), per hour pricing, automatic scaling, and in case of hardware failure, automatic recovery of VMs as per SLA.
GoGrid:2 Similar to other IaaS CSPs, GoGrid permits its clients to use a wide range of Windows and Linux images, with a scope of fixed instance sizes. GoGrid provides the services like “value-added” stacks, which are prioritized more for various applications like e-commerce, web hosting and database services.
It also provides a new feature which makes it easier for customers to combine traditional dedicated hosts with the autoscaling infrastructure facility. Customers can take the benefits of dedicated hosting combined with on-demand cloud infrastructure, which leads to the user getting the advantages of each paradigm of computation.
Rackspace Cloud:3 Rackspace provides infrastructure service that offers fixed instances of VMs in the cloud. Rackspace provides a wide scope of pre-made Linux-based images. In this cloud, RAM is measured per user-requested images. Like GoGrid, Rackspace servers combine dedicated hosts with cloud infrastructure to facilitate the customers accessing all of the features of automated cloud computing. It offers advantages such as enabling static (fixed) IP address, load balancing, and persistent storage.
2.6.2 PaaS CSPs
Aneka:4 the Aneka [16] cloud is an example of PaaS, which is a .NET-utilized platform which supports resource management and development. Each and every server acts like a cloud node and hosts Aneka container that provides the platform consisting of security and privacy, services for persistence, and control over communication. Customers request the required resources through a client to the Aneka master node, which acts as a reservation service manager and controls all the cloud nodes. Master node has the responsibility for scheduling the requests and distributing them.
The Aneka container is capable of hosting n number of other services which could be included by the developers to get the functionalities of an Aneka node.
App Engine:5 Google App Engine lets the customers run their Python program and Java applications on the cloud infrastructures provided by Google. App Engine allows the customers to deploy their