A CLOUD COMPUTING PLATFORM BASED ON P2P : A CLOUD COMPUTING PLATFORM BASED ON P2P Abstract : Abstract The Cloud computing technology has been widely applied in e-business, e-education .Cloud computing platform is a set of Scalable large-scale data server clusters, it provides computing and storage services to customers. The cloud storage is a relatively basic and widely applied service which can provide users with stable, massive data storage space. Our research shows that the architecture of current Cloud Computing System is central structured one; all the data nodes must be indexed by a master server which may become bottle neck of the system. Existing system : Existing system In these days a single server handles the multiple requests from the user. Here the server has to process the both the request from the user simultaneously, so the processing time will the high. This may leads to loss of data and packets may be delayed and corrupted. On doing this the server cannot process the query from the user in a proper manner. So the processing time gets increased. It may leads to traffic and congestion. To over come these problems we are going for the concept called “cloud computing”. In this cloud computing we are going to implement the chunk server to avoid these problems Proposed system : Proposed system New cloud storage architecture based on P2P and design a prototype system. The system based on the new architecture has better scalability and fault tolerance. A luster consists of a single master and multiple chunk servers and is accessed by multiple clients. Chunk servers store chunks on local disks and read or write chunk data specified by a chunk handle and byte range. The master maintains all file system metadata. This includes the namespace, access control information, the mapping from files to chunks, and the current locations of chunks. When a client wants to visit some data on a chunk server, it will first send a request to the Master, and the master then replies with the corresponding chunk handle and locations of the replicas. Slide 5: Software specifications:
Java1.6 or More
Java Swing – front end
Windows 98 or more.
Hard disk : 40 GB
RAM : 128mb
Processor : Pentium Modules : Modules Client :
The client application which wants to get the data from the platform.
The entity which can transfer the request or response between the Client App with the network and can lead the request to the
nearest node in the network.
The entity which is served as the data resource node and P2P node. Different with the function of pure data storage in GFS, the chunk server here has three function modules with separated interfaces. As shown in the figure above: Index Module, take charge of part of the global resource index which is assigned by DHT arithmetic such as Chord, Pastry and so on. Route Module, pass a lookup request by a next hop routing table which is also assigned by DHT. Data Module, provide the data resource stored in the local machine Architecture diagram : Architecture diagram Cloud computing: System architecture : System architecture Data flow diagram : Data flow diagram Future work : Future work we propose a new architecture of cloud computing system based on P2P protocol, which resolve the problems of bottle neck come from central structure. In the future work, we will do some optimization about the throughput of the system by the technique such as pipelining read or write. References : References Boss G, Malladi P, Quan D, Legregni L, Hall H.Cloud computing. IBM White Paper, 2007.
Ghemawat S, Gobioff H, Leung ST. The Google file system. In: Proc. of the 19th ACM Symp. On Operating Systems Principles. New York: ACM Press, 2003. 29_43.
Amazon. Amazon elastic compute cloud (AmazonEC2). 2009. http://aws.amazon.com/ec2/
Francesco Maria Aymerich, Gianni Fenu, SimoneSurcis. An Approach to a Cloud Computing Network.
Barroso LA, Dean J, Hölzle U. Web search for aplanet: The Google cluster architecture. IEEE Micro,2003,23(2):22_28.
Brin S, Page L. The anatomy of a large-scale hypertextual Web search engine. Computer Networks,1998,30(1-7):107_117.
Dean J, Ghemawat S. MapReduce: Simplified data processing on large clusters. In: Proc. of the 6th Symp.on Operating System Design and Implementation.Berkeley: USENIX Association, 2004. 137_150.
Burrows M. The chubby lock service for loosely-coupled distributed systems. In: Proc. of the 7th USENIX Symp. on Operating Systems Design and Implementation. Berkeley: USENIX Association,
Chang F, Dean J, Ghemawat S, Hsieh WC, Wallach DA, Burrows M, Chandra T, Fikes A, Gruber RE. Bigtable: A distributed storage system for structured data. In: Proc. of the 7th USENIX Symp.on Operating Systems Design and Implementation.Berkeley: USENIX Association, 2006. 205_218.