Analysis and Research of Cloud Computing System to Comparison of Several Cloud Computing Platforms

Transcription

1 Volume 1, Issue 1 ISSN: International Journal of Engineering Technology & Management Research Journal homepage: Analysis and Research of Cloud Computing System to Comparison of Anukriti Shrivastava Rajiv Gandhi Technical University, Bhopal, (M.P.), [India] anu.sys123@gmail.com Abstract Cloud computing is the development of parallel computing, distributed computing and grid computing. It has been one of the most hot research topics. Cloud Computing distribute computation task on the resource pool which consists of massive computers, accordingly, the Application systems can gain the computation strength, the storage space and software service according to its demand. Cloud computing can be viewed from two different aspects. One is about the cloud infrastructure which is the building block for the up layer cloud application. The other is of course. By means of three technical methods, cloud computing has achieved two important goals for the distributed computing: high scalability and high availability. The basic principles of cloud computing is to make the computing be assigned in a great number of distributed computers, rather than local computer or remoter server. The running of the enterprise s data center is just like Internet. With the development of parallel computing, Distributed computing, grid computing, a new computing model appeared. 1. INTRODUCTION The Cloud Computing is one kind of emerging business accounting model. It will calculate the duty to distribute on the resource pool which the massive computers constitute, enables each kind of application system according to need to gain the computation strength, the storage space and all kinds of software service. This kind of resource pool is called the cloud. The cloud is virtual computation resources that can maintain and manage itself, usually for some large -scale server cluster, including calculating server, storage server, and the broad band resources and so on. The Cloud Computing will concentrate all computation resources, and can be managed automatically through the software without intervene. This makes users will not be worried by the tedious details and concentrate his own business. This is favorable to innovation and reduces cost. The Cloud Computing is not actually the revolutionary recent development, but is the result of continuous evolution of data management technology. On the century at the end of 80's, with the development of the Internet, it starts to appear that the application massive systems solve the single question; this is the grid computation concept. It composes using the Internet the dispersible computer a hypothesized supercomputer. Each participates in the computation the computer constituting one the node, but the entire computation is composed one by the thousands of nodes the grid, provides new duty which before this is unable to complete. May say the grid International Journal of Engineering Technology & Management Research Vol 1 Issue 1 February

2 calculated first time has pushed to the IT world the resources and the service. On the century 90's, the virtualization gradually moved towards the application level by the concept, the public computation arises at the historic moment. It the computer colony which commands the grid computation took the hypothesized platform, uses the service pattern which may measure to carry on the computation. Through the public computing service, including the hardware, the software all computation resources all provides by the service provider, the customer only must needs the resources through the appropriation network or the Internet visit and defers to the actual service condition to pay expenses then. The software application pattern innovates completely which as one kind, SaaS (Software as a Service) starts the vigorous development. The user does not need to purchase the software, but is according to own actual demand, rents to the provider based on the Web software, manages the enterprise to manage the activity. SaaS while provides Internet application to the customer, also provides the software the off-line operation and the local data storage, lets software and service which the user all may use it anytime and anywhere to order. Is different with the public computation, it is not according to the consumption resources charge, but is according to the application procedure value charge which provides to the user. This kind of type service passes to through the browser the procedure the tens of thousands of users. Looked like in the user that, like this can omit the server the fixed hardware investment and the software authorization expenditure; Looked from the supplier angle, so long as like this the maintenance procedure has sufficed, can reduce the cost. 2. ABICLOUD CLOUD COMPUTING PLATFORM Abicloud is a cloud computing platform developed by Abiquo, a company locates in Barcelona Spain that is mainly focused on the development of cloud platform. It can be used to build, integrate and manage public as well as private cloud in the homogeneous environments [2]. Using Abicloud, user can easily and automatically deploy and manage the server, storage system, network, virtual devices and applications and so on. The main difference between Abicloud and other cloud computing platforms is its powerful web-based management function and its core encapsulation manner. Using the Abicloud, user can finish deploying a new service by just dragging a virtual machine with mouse. This is much easier and flexible than other cloud computing platforms that deploy new services through command lines. According to Abicloud, there is no perfect cloud platform. For each user needs his own cloud infrastructures, and every cloud provider has his own management tools, say monitor, billing and so on, so generally it is very hard to deploy a cloud platform according to user s requirement. The best way to meet the requirement of users is to build public or private cloud with homogeneous cloud computing core and extensible infrastructures. Besides, the cloud platform should also have all kinds of interfaces that support the third parties products. With all these characteristics, providers can build the cloud computing platform of their owns. Abicloud can be used to deploy and implement private cloud as well as hybrid cloud according to the cloud providers request and configuration. It can also manage EC2 according to the rules of protocol. Besides, apply the International Journal of Engineering Technology & Management Research Vol 1 Issue 1 February

3 Abicloud, a whole cloud platform based on Abicloud can be packed and redeployed at any other Abicloud platform. This is much helpful for the transformation of the working environment and will make the cloud deployment process much easier and flexible. The architecture of Abicloud is: service level protocol that permit users lease network for computing capability. Currently, Eucalyptus is compatible with EC2 from Amazon, and may support more other kinds of clients with minimum modification and extension.. Figure1 The structure of AbiCloud platform It can easily figure out that Abicloud is built based on Java, which set it irrelevant to the platform and easy to transplant. Actually, Abicloud can support many different virtual machine platforms which include vbox, VMWare, Xen, KVM and so on which make it very flexible. 3. EUCALYPTUS CLOUD PLATFORM Eucalyptus (Elastic Utility Computing Architecture for Linking Your Programs to Useful Systems) project began from California University Santa Barbara, and mainly was used to build open-source private cloud platform [8]. Now it has been run by Eucalyptus System Company. Eucalyptus is an open-source implementation of Amazon EC2 and compatible with business interfaces. It also implements virtualization depending on Linux and Xen as EC2 does. Eucalyptus is an elastic computing structure that can be used to connect the users' programs to the useful systems, it is an open-source infrastructure using clusters or workstations implementation of elastic, utility, cloud computing and a popular computing standard based on Figure 2 the resource topology structure of Eucalyptus In figure 2, node controller is a component running on the physical resources, on which all kinds of entities of virtual machine can run. It answers for the startup, check, shutdown and clearup of the virtual machines. Logic connected node controllers form a virtual cluster, all nodes belong to the same virtual cluster report to the cluster controller and are under the control and management of the cluster controller. Virtual cluster controller runs on the head node or server of the virtual cluster, is used to access private or public network. Cloud controller is the core of the manager of cloud platform, a component answering for global decision-making which is transplant to users. A Eucalyptus cloud has only one cloud controller. In Eucalyptus, client interface is the pass of communication and connection between the interior and the outside of Eucalyptus, through which users can access all kinds of resources on the cloud computing platform. 4. GOOGLE S CLOUD COMPUTING PLATFORM Google had set of exclusive Cloud Computing Platform, this Platform is first is the Google most important search International Journal of Engineering Technology & Management Research Vol 1 Issue 1 February

4 application provides the service, now already expanded to other application procedure. Google Cloud the Computing foundation construction pattern independent also is mutually close including 4 unifies in together system: The establishment above colony's Google File System, the Map Reduce programming pattern which proposed in view of the Google application procedure characteristic, distributional lock mechanism Chubby as well as the Google development model simplifies large-scale distributional database Big Table. 1) Google File System Except the performance, outside expandable, the reliability as well as the usability, the GFS design also receive the Google application load and the technical environment influence. Manifests in 4 aspects: Considers the massive nodes fully the expiration question, will need through the software fault tolerant well as the selfrecovery function integration in the system. The structure special filing system parameter, the document usual size by G byte idea, and contains the massive small documents. Full consideration application characteristic, increase d o c u m e n t s u p p l e m e n t operation, optimized in turn read-write speed. Filing system certain concrete operations no longer is transparent, needs the application procedure to assist to complete. Figure 2 has given the Google File System construction. Figure 2. Google File System architecture A GFS colony contains a host server and many block server, by many client side visit. The big document is divided the fixed size the block, the block server the block took the Linux document preservation on the local hard disk, and according to the block sentence handle and the byte scope which assigns comes the read-write block data. In order to guarantee the reliability, each block default is preserved 3 backups. Host server management filing system all Yuan data, including name space, access control, document to block mapping, block physical location and so on related information. Through the server end and the client side union design, GFS to applies the support to achieve the performance and usability most superior. GFS are, in the interior which Google application procedure itself designs has deployed many GFS colonies. Some colonies have surpass 1000 memory nodes, surpasses 300T the hard disk space, continuously frequently is being visited by the different machine on several hundred client sides. 2) Map Reduce Distributional Programming Environment Google the structure Map Reduce programming standard simplifies the distributional system the programming. The application programming personnel only must place the energy application procedure itself, but about colonies processing question, including reliability and extendibility, then hands over by the International Journal of Engineering Technology & Management Research Vol 1 Issue 1 February

5 platform processes Map Reduce to pass Map and Reduce such two simple concepts constitutes the operation basic unit, the user only must provide own Map function as well as the Reduce function then parallel processing magnanimous data. 3) Distributional Large-scale Database Management System BigTable Google the structure MapReduce programming standard simplifies the distributional system the programming. The application programming personnel only must place the energy application procedure itself, but about colony's processing question, including reliability and extendibility, then hands over by the platform processes. Map Reduce to pass Map and Reduce such two simple concepts constitutes the operation basic unit, the user only must provide own Map function as well as the Reduce function then parallel processing magnanimous data. Blue Cloud by a data central, Tivoli ProvisioningManager, IBM Tivoli Monitoring, IBM WebSphere using the server, IBM the DB2 database as well as operates the source information processing software and operates the source virtualization software to be composed together. Blue Cloud the hardware platform environment and the general x86 server cluster were similar, use the bit the way to increase the computation density. Blue Cloud the software platform characteristic mainly manifests in the virtual machine as well as regarding in the large-scale data processing software Apache Hadoop use. Hadoop is opens the source edition Google File the System software and the Map Reduce programming standard. 1) Virtualization in Blue Cloud Blue Cloud a software important characteristic is the virtualization technology use. The virtualization way may realize in the Cloud Computing in two ranks. A rank is realizes the virtualization in the hardware rank. The hardware rank virtualization may use the IBM p series server, obtains the hardware logical district LPAR. The logical district CPU resources can manage through IBM Enterprise Workload Manager. Adds on through such way in the actual use process the resource distribution strategy, can cause the corresponding resources to assign reasonably to each logical district. Virtualization other rank may obtain through the software, calculated in the platform in Blue Cloud to use the Xen virtualization software. Xen is one opens the source the virtualization software, can move other operating system above the Linux foundation, and carries on the software deployment and the operation nimbly through the virtual machine way. The virtual machine is a kind of special software, can simulate the hardware completely the execution, therefore can move the operating system in above, then can retain the one whole set movement environment semantics. Crosses the virtual machine the way, calculates in the platform in the cloud to obtain the following some merits The Cloud Computing management platform can dynamic calculate in the physical node which the platform locates needs, but does not need to stop the movement in the virtual machine platform application procedure, the advancement migration method is more nimble; Reduces the colony electrical energy consumption, will be many a load will not be the very heavy virtual machine computation node merges to the identical physical node in, thus will be able to close the idle physical node, achieved will save the electrical energy goal. International Journal of Engineering Technology & Management Research Vol 1 Issue 1 February

6 Through the virtual machine in different physical node dynamic migration, migrated the whole hypothesized movement environment, can obtain the load balanced performance which has nothing to do with the application; Is also more nimble in the deployment, namely may deploy directly the virtual machine physics calculates in the platform, but virtual machine itself has included the corresponding operating system as well as the corresponding application software, the massive virtual machine reflection will duplicate the correspondence directly the physical node then. 2) Memory Structure in Blue Cloud The Blue Cloud memory architecture regarding the Cloud Computing also is extremely important, regardless of is the operating system, the service routine or the user application procedure data all preserves in the memory system. Blue Cloud the memory architecture contains is similar to the Google File System colony filing system as well as based on the block equipment memory regional network SAN system. On Blue Cloud, SAN systematic and the distributed file system (e.g. Google File System) is not the system which opposes mutually, SAN provides is the block equipment connection, needs to construct the filing system in this foundation, can use by the upper formation application procedure. But Google File System happen to be a distributional filing system, can establish above SAN. Both all can provide the reliability, the extendibility, how as for uses also needs to calculate in the platform based on the establishment in the cloud the application procedure to decide, this also manifested has calculated the relations which the platform and the upper formation application cooperated mutually. 5. COMPARISON OF CLOUD PLATFORMS Currently, there are kinds of cloud computing platforms, each has its own characteristics and advantages. To better understand these platforms, we analyze in detail and give a comparison from different implementation aspects. The characteristics and implementation of these platforms are summarized as table 1 shows. From table 1, it can figure out that though the implementation of these cloud platforms is quite different, there are much common between them. For example, they are all scalable, all provide IaaS, all support dynamic deployment of the platform, all support Xen virtualization technology, and all support Linux operation system and the development of application with Java. However, there are also many differences, say their network interfaces, structure and reliability and so on. Generally, each cloud platform has its own advantages over others. For example, from the point of cloud platform deployment, Abicloud stands out. As this cloud platform can be deployed with mouse under graphic user interfaces compared others with command line. This will be much simple to users and decrease the effort of the platform deployment. From the point of reliability, Open Nebula is more mature. For it has considered rollback and fault tolerance mechanisms in the cloud implementation while others do not. Table 1 The comparison of several cloud computing platforms International Journal of Engineering Technology & Management Research Vol 1 Issue 1 February

7 6. CONCLUSIONS Cloud computing is a new technology widely studied in recent years. Now there are many cloud platforms both in industry and in academic circle. How to understand and use these platforms is a big issue. Focused on the aspects such as the architectures, characteristics, application and so on, a detailed comparison has been presented in this paper. From the anlysis and summarization, users can better understand the characteristics and better choose of cloud computing platforms according to the cloud types, interfaces, compatibility, implementation, deployment requirement, and development support and so on. Though each cloud computing platform has its own strength, one thing should be noticed is that no matter what kind of platform there is lots unsolved issues. For example, continuously high availability, dealt mechanisms of cluster failure in cloud environment, consistency guaranty, synchronization in different clusters in cloud platform, interoperation and standardization, the security of cloud platform and data in transmission and so on are all among the issue to be better solved. REFERENCES : [1] Elastic Compute Cloud Amazon Web Services. [2] [3] Microsoft Azure. [4] [5] Google App Engine. [6] [7] Sims K. IBM introduces ready-touse cloud computing collaboration [8] services get clients started with cloud computing [9] en/pressrelease/22613.wss. [10] Citrix systems, citrix XenServer: Efficient virtual server software. [11] XenSource Company. [12] Global Cloud computing test bed.amazon Simple Storage S e r v i c [ U R L ]. h t t p : / / aws.amazon.com/s3/,access on Sep International Journal of Engineering Technology & Management Research Vol 1 Issue 1 February