IBM Power Systems 775 Manual page 91

GPFS plays a key role in the shared storage configuration for Power 775 clusters. Virtually all
large-scale systems are connected to disk over HFI via GPFS Network Shared Disk (NSD)
servers, which are referred GPFS I/O nodes or Storage nodes in Power 775 terminology. The
system interconnect features higher performance and is far more scalable than traditional
storage fabrics, and is RDMA capable.
GPFS includes a Native RAID function that is used to manage the disks in the disk
enclosures. In particular, the disk hospital function is queried regularly to ascertain the health
of the disk subsystem. This function is not always necessary because disk problems that
require service are reported to the HMC serviceable events and to TEAL.
For more information about GPFS, see Table 1-6 on page 62.
GPFS Native RAID
GPFS Native RAID is a software implementation of storage RAID technologies within GPFS.
By using conventional dual-ported disks in a Just-a-Bunch-Of-Disks (JBOD) configuration,
GPFS Native RAID implements sophisticated data placement and error correction algorithms
to deliver high levels of storage reliability, availability, and performance. Standard GPFS file
systems are created from the NSDs defined through GPFS Native RAID.
This section describes the basic concepts, advantages, and motivations behind GPFS Native
RAID: redundancy codes, end-to-end checksums, data declustering, and administrator
configuration, including recovery groups, declustered arrays, virtual disks, and virtual disk
NSDs.
Overview
GPFS Native RAID integrates the functionality of an advanced storage controller into the
GPFS NSD server. Unlike an external storage controller, in which configuration, LUN
definition, and maintenance are beyond the control of GPFS, GPFS Native RAID takes
ownership of a JBOD array to directly match LUN definition, caching, and disk behavior to
GPFS file system requirements.
Sophisticated data placement and error correction algorithms deliver high levels of storage
reliability, availability, serviceability, and performance. GPFS Native RAID provides a variation
of the GPFS NSD called a
the VDisk NSDs of a file system by using the conventional NSD protocol.
The GPFS Native RAID includes the following features:
Software RAID: GPFS Native RAID runs on standard AIX disks in a dual-ported JBOD
array, which does not require external RAID storage controllers or other custom hardware
RAID acceleration.
Declustering: GPFS Native RAID distributes client data, redundancy information, and
spare space uniformly across all disks of a JBOD. This distribution reduces the rebuild
(disk failure recovery process) overhead that is compared to conventional RAID.
Checksum: An end-to-end data integrity check (by using checksums and version
numbers) is maintained between the disk surface and NSD clients. The checksum
algorithm uses version numbers to detect silent data corruption and lost disk writes.
Data redundancy: GPFS Native RAID supports highly reliable two-fault tolerant and
three-fault-tolerant Reed-Solomon-based parity codes and three-way and four-way
replication.
Large cache: A large cache improves read and write performance, particularly for small
I/O operations.
virtual disk
, or VDisk. Standard NSD clients transparently access
Chapter 1. Understanding the IBM Power Systems 775 Cluster
77