Metadata server

A metadata server is a software server that performs metadata, administrative, and storage-management services and provides clients with shared, coherent access to shared storage (or global namespace). The metadata servers are clustered for scalability and availability, and are often referred to as a cluster. In the cluster, there is one master metadata server and one or more subordinate metadata servers, each running on a separate storage engine. Additional metadata servers can be added, as required, when the workload grows.

All of the metadata servers, including the master metadata server, share the workload of the global namespace. Each is responsible for providing metadata and locks to clients for specific filesets assigned to them. They know which filesets belong to which metadata server, and when contacted by a client, can direct the client to the appropriate metadata server. They also manage distributed locks to ensure the integrity of all of the data within the global namespace.

In addition to providing metadata to clients and managing locks, metadata servers perform a wide variety of other tasks. They process requests to create and manage filesets, storage pools, volumes, and policies; enforce the policies to place files in appropriate storage pools; and send alerts when any threshold established for the filesets and storage pools are exceeded.

Parent topic: Concepts

Related concepts
Policies and rules
SNMP

Related tasks
Starting a metadata server

Terms of use | Feedback
(C) Copyright IBM Corporation 2003, 2004. All Rights Reserved.