Implementation of SYNC_NV bit

Gerry Houlder gerry.houlder at seagate.com
Fri Jul 23 14:56:31 PDT 2010


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1007234_f.htm">HTML-formatted message</a>

For read and write commands, the question is not so important because the
choice only applies to the data transferred during that command and
therefore will not cause such a large performance difference compared to a
SYNCHRONIZE CACHE command that transfers no data but may cause the movement
of several Gbytes of data to "media". It is important from a standards
consistency point of view, however. If the SYNC_NV definition were changed,
the FUA_NV should change with it.
It is also getting harder to decide what is "media" and what is "cache". For
instance, a hybrid drive may keep many Gbytes of data stored in non-volatile
flash memory (for faster performance). Does this count as cache or as media,
|from the point of view of how the SYNCHRONIZE CACHE command uses those
terms? Perhaps all non-volatile memory could be treated as "media" and not
need to be synchronized (except as a background task).
The same question should apply to the FUA vs. FUA_NV bits in the READ and
WRITE commands.
 ---
Rob Elliott    HP ISS Platform Technology - Server Storage
 *From:* owner-t10 at t10.org [mailto:owner-t10 at t10.org] *On Behalf Of *Gerry
Houlder
*Sent:* Monday, 19 July, 2010 4:21 PM
*To:* T10 Reflector
*Subject:* Implementation of SYNC_NV bit
My company has been reviewing the definition of the SYNC_NV bit in the
SYNCHRONIZE CACHE command.



More information about the T10 mailing list