Table of Contents
Advertisement
All CNFS nodes export the same file systems to the NFS clients. When one of
the CNFS nodes fails, the NFS serving load moves from the failing node to
another node in the CNFS cluster. Failover is done by using recovery groups
to help choose the preferred node for takeover.
GPFS Multi-cluster
GPFS mounts GPFS file systems that are owned by other clusters in a
network of sufficient bandwidth. Each site in the network is managed as a
separate cluster, while allowing shared file system access. GPFS also allows
users shared access to files in the cluster where the file system was created,
or other GPFS clusters.
The cluster that owns the file system is responsible for administering the file
system and granting access to other clusters on a per cluster basis. After
access to a particular file system is granted to nodes in another GPFS cluster,
they can mount the file system and perform data operations as though the file
system were locally owned.
IBM GPFS Active File Management (AFM)
Active file management (AFM) is a scalable, high-performance file system
caching layer that is integrated with the GPFS cluster file system. By using
AFM, you can create associations from a local GPFS cluster to a remote
cluster or storage, and define the location and flow of file data to automate the
management of the data. You can implement a single namespace view
across sites around the world.
AFM masks wide area network latencies and outages by using GPFS to
cache massive data sets, which allows data access and modifications even
when remote storage cluster is unavailable. In addition, AFM performs
updates to the remote cluster asynchronously, with which applications can
continue operating while not being constrained by limited outgoing network
bandwidth.
The AFM implementation uses the inherent scalability of GPFS to provide a
multinode, consistent cache of data at a home cluster. By integrating with the
file system, AFM provides a Portable Operating System Interface for UNIX
(POSIX)-compliant interface, which makes the cache not apparent to
applications. AFM is easy to deploy because it relies on open standards for
high-performance file serving and does not require any proprietary hardware
or software to be installed at the home cluster.
The following features that use IBM GPFS capabilities are available through a
different licensing:
IBM GPFS File Placement Optimizer (FPO)
IBM GPFS FPO extends GPFS for a new class of data-intensive applications,
which are commonly referred to as big data applications.
Chapter 8. Software stack
221
Advertisement
Table of Contents
