Scribd Logo
Upload

Welcome to Scribd!

  • 61 views

    Seminar Report Format

    Uploaded by

    Michael Kelley
    This document provides an abstract for a seminar presentation on data replication techniques in a grid computing environment. The abstract discusses how fragmented replicas, which store only partial contents of files locally, can improve storage efficiency. It also describes how downloading data from multiple replicated locations in parallel can improve data retrieval performance by allowing for parallel downloading and establishing connections from multiple servers to a client. The document includes sections on acknowledgements, table of contents, introduction, software and technologies used, implementation details, conclusions, and future work.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as DOC, PDF, TXT or read online from Scribd

    Seminar Report Format

    Uploaded by

    Michael Kelley
    61 views8 pages
    This document provides an abstract for a seminar presentation on data replication techniques in a grid computing environment. The abstract discusses how fragmented replicas, which store only partial contents of files locally, can improve storage efficiency. It also describes how downloading data from multiple replicated locations in parallel can improve data retrieval performance by allowing for parallel downloading and establishing connections from multiple servers to a client. The document includes sections on acknowledgements, table of contents, introduction, software and technologies used, implementation details, conclusions, and future work.

    Original Description:

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    DOC, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides an abstract for a seminar presentation on data replication techniques in a grid computing environment. The abstract discusses how fragmented replicas, which store only partial contents of files locally, can improve storage efficiency. It also describes how downloading data from multiple replicated locations in parallel can improve data retrieval performance by allowing for parallel downloading and establishing connections from multiple servers to a client. The document includes sections on acknowledgements, table of contents, introduction, software and technologies used, implementation details, conclusions, and future work.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as DOC, PDF, TXT or read online from Scribd
    Download as doc, pdf, or txt
    61 views8 pages

    Seminar Report Format

    Uploaded by

    Michael Kelley
    This document provides an abstract for a seminar presentation on data replication techniques in a grid computing environment. The abstract discusses how fragmented replicas, which store only partial contents of files locally, can improve storage efficiency. It also describes how downloading data from multiple replicated locations in parallel can improve data retrieval performance by allowing for parallel downloading and establishing connections from multiple servers to a client. The document includes sections on acknowledgements, table of contents, introduction, software and technologies used, implementation details, conclusions, and future work.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as DOC, PDF, TXT or read online from Scribd
    Download as doc, pdf, or txt
    of 8

    A SEMINAR ON

    SEMINAR TOPIC

    DEPARTMENT OF ELECTRONICS ENGINEERING

    BY
    STUDENT NAME (ROLL NO)

    FACULTY NAME

    SUPERVISOR

    JSS ACADEMY OF TECHNICAL EDUCATION - NOIDA MONTH, YEAR

    ii

    ABSTRACT
    Data Grids support data-intensive applications in wide area Grid systems. Replication is a commonly used technique in a distributed environment. The motivation of replication is that replication can improve data availability, reducing bandwidth consumption, data access latency, and load balancing. Usually a complete file is copied to many Grid sites for local access. However, local application at a site may only need parts of a replica. Therefore, to use the storage systems efficiently, it is necessary for a Grid site to store only parts of a replica. Fragmented replicas are proposed for this situation. That is, when doing replication, a site can store only some partial contents needed locally. It can greatly save the storage space wasted in storing unused data.

    Multiple replicas can improve the data retrieval performances in a Grid environment by allowing parallel downloading of the desired data sets. However, since network conditions and server speeds differ, how to schedule the downloading job efficiently is very important. Replicating popular content in multiple servers is a widely used technique. Using the prediction techniques several co-allocation mechanisms are developed to establish connections between servers and a client. There is only one stream in a typical Internet download between a server and a client. This may suffer

    iii

    from some problems when downloading large datasets in a Data Grid environment. The bandwidth achievable is limited by several bottlenecks. For example, one is the congestion in the link connecting the server to the client and to the Internet. The other is the outputting bandwidth of the servers network connection. It may involve the server CPU speed, disk storage speed, etc. One way to improve download speed is downloading data from multiple locations in parallel in order to minimize the problems mentioned above.

    iv

    ACKNOWLEDGEMENT

    I owe my sincere gratitude to my supervisor, Mrs. ________ (<Desingation>, < Department of ____>, JSSATE-NOIDA) who has constantly given me the encouragement, technical guidance and moral support throughout my thesis work.

    I would like to extend my heart felt thanks and lifelong indebtedness to Prof. Dinesh Chandra (HOD, Department of EC, JSSATE-NOIDA) for his _________________________________________________________

    STUDENT NAME

    TABLE OF CONTENTS
    CHAPTER NO. TITLE PAGE NO. ABSTRACT iii LIST OF FIGURES x LIST OF ABBREVIATIONS xi INTRODUCTION TO GRID COMPUTING 1 1.1 OVERVIEW 1 1.1.1 Exploiting underutilized resources 2 1.1.2 Parallel CPU capacity 3 1.2 GRID ARCHITECTURE (SOA) 4 1.2.1 Overview 5 1.2.2 Requirements for a SOA 8 1.2.3 Web services approach to a SOA 9 1.3 OPEN GRID SERVICE ARCHITECTURE (OGSA) 11 1.3.1 Open Grid Service Interface (OGSI) 12 SOFTWARES AND TECHNOLOGIES 13 2.1 ALCHEMI.NET (GRID MIDDLEWARE) 13 2.1.1 Architecture 13 2.1.2 Components 14 2.1.3 Grid Application Lifecycle 19 2.1.4 ALCHEMI API: Grid Thread Programming Model 20 2.1.5 Grid Job Model 21 2.1.6 Overview of Grid Application Development with Alchemi 22 2.2 .NET REMOTING 25 2.2.1 Copies vs. References 26 2.2.2 Remoting Architecture Simplified 27 2.2.3 Comple Remoting System Design 28 2.2.4 Building a Basic .NET Remoting Application 29 2.3 DATA REPLICATION 35 2.3.1 Components of the Replication Process 36 2.3.2 Types of Replication 38 IMPLEMENTATION 43 3.1 LITERATURE REVIEW 43 3.2 GRID SETUP 46 3.3 PROBLEM STATEMENT 50 3.4 RESULTS 52 3.6 CONCLUSION 57 3.7 FUTURE WORK 57 REFERENCES 58

    vi

    3.5 CONCLUSION In the Grid technology, where non-trusted users and heterogeneous systems with unpredictable network connections and varying speeds are interconnected, it is first of all, necessary to ensure that all the threads are executed. Then is the issue of improvising the speed. In the context of the parallel downloading of data from a number of replicated sites, the suggested algorithm would work well and ensure the completion of all threads. But, since each thread is allocated to individual executors, one at a time, some time is elapsed between the scheduling of each thread. Hence, the algorithm is suitable in situations where, the size of the data block is large enough that

    vii

    the time required for downloading a data block is comparatively much higher than this elapsed time. This problem can also be overcome by allocating a group of data blocks instead of a single block to each server. The failed threads are to be re-scheduled in the end. 3.6 FUTURE WORK There is a great scope for this project in the area of E-Learning, where huge amount of data and study material exists. Further research regarding implementation in this particular field needs to be carried out. The experiments are to be repeated for different data sizes and also the same experiment needs to be carried out on the Cross-Platform Manager, using heterogeneous systems over the internet, and using GRIDFTP instead of the FTP protocol.

    REFERENCES

    1. Akshay Luther, Rajkumar Buyya, Rajiv Ranjan, and Srikumar Venugopal;Alchemi: A .NET-based Grid Computing Framework and its Integration into Global Grids ; Grid Computing and Distributed Systems (GRIDS) Laboratory, Department of Computer Science and Software Engineering, The University of Melbourne, Australia 2. Ian Foster, Carl Kesselman, Steven Tuecke ; The Anatomy of the Grid : Enabling Scalable Virtual Organizations ; International Journal on Supercomputer Applications, 2001 3. Luis Ferreira, Viktors Berstis, Jonathan Armstrong ; Introduction to Grid Computing with Globus ; IBM.COM/Red Books; IBM Corporation

    BIBILIOGRAPHY

    viii

    1. Christelle Gueret et.al; Building University Timetables using Constraint Logic Programming; Proceedings of first international conference on practice and theory of automated timetabling; Vol-1153; Year:1995; Pages:130-145; ISBN:3-540-61794-9; Publisher: Springer Verlag 2. Harikleia Frangouli et.al; UTSE: Construction of optimum timetables for university courses- A CLP based approach; Proceedings of the conference on the practical applications of PROLOG; Pages-225-243; Year:1995; Publisher Alinmead Software Ltd.

    WEB RESOURCES

    1. http://support.microsoft.com/kb/871044 ; How to use named pipes for


    interprocess communication in Visual Basic .NET or in Visual Basic 2005 2. http://www.codeproject.com/KB/vb/Remoting_HelloWorld.aspx ; A sample Remoting HelloWorld codeproject

    You might also like