Administrator Guide

Linear volumes
Linear volumes make use of a method of storing user data in sequential, fully allocated physical blocks. Mapping between the logical data
presented to hosts and the physical location where it is stored is fixed, or static.
About volume cache options
You can set options that optimize reads and writes performed for each volume. It is recommended that you use the default settings.
Using write-back or write-through caching
CAUTION: Only disable write-back caching if you fully understand how the host operating system, application, and
adapter move data. Used incorrectly, write-back caching can hinder system performance.
When modifying a volume you can change its write-back cache setting. Write-back is a cache-writing strategy in which the controller
receives the data to be written to disks, stores it in the memory buffer, and immediately sends the host operating system a signal that the
write operation is complete, without waiting until the data is actually written to the disk. Write-back cache mirrors all of the data from one
controller module cache to the other. Write-back cache improves the performance of write operations and the throughput of the
controller.
When write-back cache is disabled, write-through becomes the cache-writing strategy. Using write-through cache, the controller writes
the data to the disks before signaling the host operating system that the process is complete. Write-through cache has lower write
throughput performance than write-back, but it is the safer strategy, with minimum risk of data loss on power failure. However, write-
through cache does not mirror the write data because the data is written to the disk before posting command completion and mirroring is
not required. You can set conditions that cause the controller to change from write-back caching to write-through caching. For more
information, see Changing system cache settings.
In both caching strategies, active-active failover of the controllers is enabled.
You can enable and disable the write-back cache for each volume. By default, volume write-back cache is enabled. Because controller
cache is backed by supercapacitor technology, if the system loses power, data is not lost. For most applications, this is the preferred
setting.
NOTE: The best practice for a fault-tolerant configuration is to use write-back caching.
Cache optimization mode
CAUTION:
Changing the cache optimization setting while I/O is active can cause data corruption or loss. Before
changing this setting, quiesce I/O from all initiators.
You can also change the optimization mode.
• Standard. This controller cache mode of operation is optimized for sequential and random I/O and is the optimization of choice for
most workloads. In this mode, the cache is kept coherent with the partner controller. This mode gives you high performance and high
redundancy. This is the default.
• No-mirror. In this mode of operation, the controller cache performs the same as the standard mode with the exception that the cache
metadata is not mirrored to the partner. While this improves the response time of write I/O, it comes at the cost of redundancy. If this
option is used, the user can expect higher write performance but is exposed to data loss if a controller fails.
Optimizing read-ahead caching
NOTE:
Only change read-ahead cache settings if you fully understand how the host operating system, application, and
adapter move data so that you can adjust the settings accordingly.
You can optimize a volume for sequential reads or streaming data by changing its read-ahead cache settings.
You can change the amount of data read in advance. Increasing the read-ahead cache size can greatly improve performance for multiple
sequential read streams.
• The Adaptive option works well for most applications: it enables adaptive read-ahead, which allows the controller to dynamically
calculate the optimum read-ahead size for the current workload.
• The Stripe option sets the read-ahead size to one stripe. The controllers treat NRAID and RAID-1 disk groups internally as if they have
a stripe size of 512 KB, even though they are not striped.
• Specific size options let you select an amount of data for all accesses.
22
Getting started