Replica Management

1
Replica Management

• Kelly Clynes

2
Agenda

• Grid Computing
• Globus Toolkit
• What is Replica Management
• Replica Management in Globus
• Replica Management Catalog
• Metadata Catalog
• Replica Catalog
• Replica Manager

3
Grid Computing

• Emerging computational and networking
  infrastructure
• Software to divide and distribute pieces of a
  program to as many as several thousand computers

4
Globus Toolkit

• Technology for the grid
• Open source toolkit for building computing grids
• Developed by the Globus Alliance.

5
Replica Management

• Created because new storage locations offers
  better performance or availability for accesses
  to or from a particular location
• Registers new copies in a Replica Catalog
• Selects the best replica for access based on
  storage and the network
• Create or delete copies of file instances, or
  replicas, within specified storage systems

6
Replica Management

• Controls where and when copies are created, and
  provides information about where copies are
  located
• Possible to create hierarchies of replica
  catalogs to impose a directory-like structure on
  related logical collections
• Allows used and applications to query the catalog
  to find all existing copies of a particular file
  or collection of files

7
Replica Management in Globus

• Integrates the Globus Replica Catalog (for
  keeping track of replicated files)
• GridFTP (for moving data) and provides replica
  management capabilities for data grids.
• Two main catalogs
• The Metadata Catalog
• The Replica Catalog

8
Metadata Catalog

• Describes the content that the data withholds.
• Entries in the catalog correspond to logical
  files and possibly collections of logical files
• Must support a variety of different metadata
  catalogs as well as produce logical names for the
  collections and file

9
The Replica Catalog

• Lowest layer of the Globus replica management
  architecture
• Provides mappings between the logical names for
  collections and files with the copies of the
  objects on the physical storage systems.
• Three types of entries
• logical collections
• locations
• logical files

10
Replica Catalog API

• Implemented as well as a command line tool
• Perform low-level manipulation operations for
  replica catalog
• Creating
• Deleting
• Modifying

11
Replica Manager

• Replica System Manager
• Manages the copy of a file
• Maintains a mapping between logical names for
  files and collections and for one or more
  physical locations

12
Summary

• Grid Computing
• Globus Toolkit
• Replica Management
• Replica Management and Globus
• Replica Manager

13
Replica Management

• Questions?