Technical data

GL–20 Glossary
write-back caching A cache management method used to decrease the subsystem’s response
time to write requests by allowing the controller to declare the write
operation “complete” as soon as the data reaches its cache memory. The
controller performs the slower operation of writing the data to the disk
drives at a later time.
write-through
caching
A cache management method used to decrease the subsystem’s response
time to a read. This method allows the controller to satisfy the request from
the cache memory rather than from the disk drives.
write hole The period of time in a RAID level 1 or RAID level 5 write operation when
an opportunity emerges for undetectable RAIDset data corruption. Write
holes occur under conditions such as power outages, where the writing of
multiple members can be abruptly interrupted. A battery backed-up cache
design eliminates the write hole because data is preserved in cache and
unsuccessful write operations can be retried.
write-through cache A cache management technique for retaining host write requests in read
cache. When the host requests a write operation, the controller writes data
directly to the storage device. This technique allows the controller to
complete some read requests from the cache, greatly improving the
response time to retrieve data. The operation is complete only after the data
to be written is received by the target storage device.
This cache management method may update, invalidate, or delete data from
the cache memory accordingly, to ensure that the cache contains the most
current data.