Configuration and Tuning GPFS for Digital Media Environments
Published 06 October 2005
Authors: Octavian Lascu
In today’s environment, when people demand access to more and more information, media companies are implementing new solutions for their environments, focusing on performance, availability and cost.
Because GPFS has its historical roots in multimedia, the adoption of this product for digital media environments comes as a natural evolution.
This IBM Technote focuses on (but is not limited to) the newest GPFS functionality and its on demand characteristics for digital media environments:
- Digital media information flow and requirements analyzed and matched with GPFS functionality
- Interoperability of a GPFS cluster and file sharing applications for Windows clients
The IBM Redpaper for this Technote presents a nine-step approach for configuring a GPFS cluster, and also presents various tools and techniques that make a systems administrator's life easier for GPFS implementation and day-to-day management tasks.
GPFS replication (disk failure groups)
General Parallel File System (GPFS) is a journaling file system that logs all metadata activity. Several protection mechanisms are used to maintain the file system online, or mounted:
- Data and metadata replication
- Log file replication
- Disk failure groups
The disk failure groups are used for keeping the file system, not the cluster, online, based on maintaining a quorum of file system descriptors (disks) available. If you are familiar with AIX logical volume manager, the information about a volume group is stored on the Volume Group Descriptor Area (VGDA). A quorum of VGDAs must be available for the VG to be either brought online or to remain varied on.
GPFS reserves a storage area on every disk which is part of a file system for the file system descriptor (FSDesc). However, by default writes the FSDesc on three of the disks which are part of that file system, if available.
This behavior poses the problem that in the event of multiple-disk failure, the file system can be unmounted due to the lack of an FSDesc quorum, even though the total number of disks still available for that file system is larger that 50%+1.
To solve this problem, GPFS uses disk failure groups. The characteristics of the failure groups are that they:
- Have a number identifying the failure group to which a disk belongs
- Are used by GPFS during metadata and data placement on the disks of a file system
- Ensure that no two replicas of the same block will become unavailable due to a single failure
- Are set by GPFS by default
- Can also be defined either at NSD creation time (mmcrnsd/mmcrvsd commands) or later (mmchdisk command)
The syntax for the disk descriptor file is:
It is important to set failure groups correctly to have proper and effective file system replication. This is true for metadata and data.
This material has not been submitted to any formal IBM test and is published AS IS. It has not been the subject of rigorous review. IBM assumes no responsibility for its accuracy or completeness. The use of this information or the implementation of any of these techniques is a client responsibility and depends upon the client's ability to evaluate and integrate them into the client's operational environment.
Follow IBM Redbooks
Follow IBM Redbooks