Presenter: Rajendra Singh

1
Replica Management in Grid

• Presenter Rajendra Singh
• cs 783 A Survey on Replica Management in Grid
• Instructor Prof. Sajid Hussain

2
Outline

• The Grid and Data Grid
• Motivation
• Survey area
• References

3
What is Grid?

• The term The Grid came into existence in the
  middle 1990s.
• Denotes the construction of a Nation-Wide
  (Global) computing infrastructure
• in the form of a very large scale distributed
  computing system built by connecting
  geographically distributed machines and other
  network resources available at various
  institutions and organizations.

4
Data Grid and Motivation

• Aim is to enable access to geographically
  distributed computing and storage facilities.
• Huge amount of data in order of 100s of
  Gigabytes is generated per day
• High Energy Physics,
  
• Biology and Medical Image processing,
• Earth Observations, etc.

some application areas
5
Motivation

• Researchers who want to use data are
  geographically distributed.
• So is the data.
• Not feasible to access data at single location
• data access latency,
• applications suffer in terms of performance,
  starve for data,
• not a good utilization of resources available
• neither computing, nor storage (in such a
  distributed environment).

6
Replica Management Survey

• Replicas of data can be stored at various
  distributed sites.
• reduces data access latency,
• application performance speed-up,
• Issues,
• granularity of replication,
• propagation of updates (mechanisms required),
• consistency in metadata and maintenance of
  metadata catalog, etc.

7
References

• A. Chervenak, I. Foster, C. Kesselman, C.
  Salisbury, S. Tuecke. The Data Grid Towards an
  Architecture for the Distributed Management and
  Analysis of Large Scientific Datasets. Journal of
  Network and Computer Applications, 23187-200,
  2001.
• B. Allcock, J. Bester, J. Bresnahan, A. L.
  Chervenak, I. Foster, C. Kesselman, S. Meder, V.
  Nefedova, D. Quesnal, S. Tuecke. Data Management
  and Transfer in High Performance Computational
  Grid Environments. Parallel Computing Journal,
  Vol. 28 (5), May 2002, pp. 749-771.
• Further reading
• http//www.globus.org/research/papers.html

8

• Thanks