Intel RAID Controllers - Best Practices white paper

Intel® RAID Controllers Best Practices White Paper
Revision 1.0
8
5. When to Use a RAID Controller Battery
A RAID controller battery should be used whenever virtual drive write-back cache is enabled
and data is mission critical.
Cache-to-cache I/O is much faster than any other type of I/O operation occurring on the data
bus. It is faster to write data to the RAID adapter’s cache memory than it is to write it directly to
a storage device because the time required to spin target data under a read or write head is
longer than the time required to perform the read or write to a memory device.
If the RAID Controller’s write-back cache option is enabled, data is first written to the cache
memory and the write is acknowledged, and then the RAID controller writes the cached data to
the storage device when it is available to service the I/O request. However, this method of
writing data first to cache memory, acknowledging the write as complete, and then completing
the write when the drive is available carries inherent risk. Cached data on the RAID controller
can be lost if the AC power fails before the cached data is written to the storage device. The
Smart Battery mitigates this risk by providing battery power to the RAID controller memory and
holding the data in the RAID cache memory until power is restored. The battery can hold data
in the RAID controller’s memory for up to 72 hours.
The Smart Battery accomplishes all of this by monitoring the voltage level of the DRAM modules
on the RAID controller. If the voltage drops below a defined level, the Smart Battery switches
the memory power source from the RAID controller to the battery pack. The battery pack
provides power for the memory until the voltage returns to an acceptable level, at which time the
Smart Battery circuit board switches the power source back to the RAID controller. Cached data
is then written to the storage device just as though the power loss had never occurred. The
Smart Battery provides additional fault tolerance even when used with a UPS, which does not
prevent a system power supply failure or other system internal power failure.
6. Why Use a UPS?
An uninterruptible power supply (UPS) is a battery-based system power supply that helps
protect electronic equipment from an unexpected loss of power. A UPS is highly recommended
to protect data in mission critical configurations. Computers and accessories can suffer damage
during a power outage or experience a loss of data that is in transit during the power outage.
There is no way to provide a battery backup of data that is temporarily stored in the hard disk
cache but has not been written to disk. A power outage could corrupt the data on a server or
make data unavailable to users. A UPS can reduce the chance that a power outage could
corrupt data on a server. Although the addition of UPS is not a guarantee that data cannot be
lost, it does add additional security.