Fault Tolerance; Multipathing - Lenovo ThinkServer RD650 User Manual

12Gb/s MegaRAID SAS Software User Guide
March 2014
a part of a drive group

parts of more than one drive group

a combination of any two of these conditions

2.1.3

Fault Tolerance

Fault tolerance is the capability of the subsystem to undergo a drive failure or failures without compromising data
integrity, and processing capability. The RAID controller provides this support through redundant drive groups in
RAID levels 1, 5, 6, 10, 50, and 60. The system can still work properly even with drive failure in a drive group, though
performance can be degraded to some extent.
In a span of RAID 1 drive groups, each RAID 1 drive group has two drives and can tolerate one drive failure. The span of
RAID 1 drive groups can contain up to 32 drives, and tolerate up to 16 drive failures—one in each drive group. A
RAID 5 drive group can tolerate one drive failure in each RAID 5 drive group. A RAID 6 drive group can tolerate up to
two drive failures.
Each spanned RAID 10 virtual drive can tolerate multiple drive failures, as long as each failure is in a separate drive
group. A RAID 50 virtual drive can tolerate two drive failures, as long as each failure is in a separate drive group.
RAID 60 drive groups can tolerate up to two drive failures in each drive group.
Fault tolerance is often associated with system availability because it allows the system to be available during the
failures. However, fault tolerance means that it is also important for the system to be available during the repair of the
problem.
A hot spare is an unused drive. You can use a hot spare to rebuild the data and re-establish redundancy in case of a
disk failure in a redundant RAID drive group. After the hot spare is automatically moved into the RAID drive group, the
data is automatically rebuilt on the hot spare drive. The RAID drive group continues to handle requests while the
rebuild occurs.
Auto-rebuild allows a failed drive to be replaced and the data automatically rebuilt by hot-swapping the drive in the
same drive bay. The RAID drive group continues to handle requests while the rebuild occurs.
2.1.3.1

Multipathing

The firmware provides support for detecting and using multiple paths from the RAID controllers to the SAS devices
that are in enclosures. Devices connected to enclosures have multiple paths to them. With redundant paths to the
same port of a device, if one path fails, another path can be used to communicate between the controller and the
device. Using multiple paths with load balancing, instead of a single path, can increase reliability through redundancy.
Applications show the enclosures and the drives connected to the enclosures. The firmware dynamically recognizes
new enclosures added to a configuration along with their contents (new drives). In addition, the firmware dynamically
adds the enclosure and its contents to the management entity currently in use.
Multipathing provides the following features:
Support for failover, in the event of path failure

Auto-discovery of new or restored paths while the system is online, and reversion to system load-balancing policy

Measurable bandwidth improvement to the multi-path device

Support for changing the load-balancing path while the system is online

The firmware determines whether enclosure modules (ESMs) are part of the same enclosure. When a new enclosure
module is added (allowing multi-path) or removed (going single path), an Asynchronous Event Notification (AEN) is
generated. AENs about drives contain correct information about the enclosure, when the drives are connected by
multiple paths. The enclosure module detects partner ESMs and issues events appropriately.
NOTE RAID level 0 is not fault tolerant. If a drive in a RAID 0 drive group fails,
the entire virtual drive (all drives associated with the virtual drive) fails.
LSI Corporation
- 19 -
Chapter 2: Introduction to RAID
Components and Features