Power Condition Mode Select interpretation of intent

Gerry.Houlder at seagate.com Gerry.Houlder at seagate.com
Mon Aug 25 11:19:51 PDT 2008


* From the T10 Reflector (t10 at t10.org), posted by:
* Gerry.Houlder at seagate.com
*
There is no intent to change the existing behavior of mode page parameters.
When a timer value is set (including set to zero) that value is retained by
the SCSI target device until changed by another MODE SELECT command or a
reset (i.e., if the saved value were different than the current value).
Therefore setting the timer value to zero is not a one time event, it will
cause the target device to go to that power condition immediately after
completing each command. In my opinion, this is not a good choice; there
should be some delay before invoking a lower power condition to keep the
target device from power cycling too frequently and actually causing higher
power consumption.
Direct Access devices use the START STOP command to invoke the "go
immediate" function without modifying the mode page settings. See 08-184
for enhancements on this. There apparently is not a command that does this
function for other device types. This is the reason Rob Elliott's power
management proposal (08-126) creates a new "power managment" command that
could apply to all device types. Devices other than direct access type
should be interested in this feature.
	     Kevin D Butt						   
	     <kdbutt at us.ibm.co						   
	     m> 							To 
	     Sent by:		       t10 at t10.org			   
	     owner-t10 at t10.org						cc 
	     No Phone Info	       Hirokazu Nakayama <HNAK at jp.ibm.com> 
	     Available						   Subject 
				       Power Condition Mode Select	   
				       interpretation of intent 	   
	     08/25/2008 10:58						   
	     AM 							   
I have a question of the intent of the Power Condition mode parameters.
The following text is from SPC-4r16.
<<
7.4.12 Power Condition mode page
The Power Condition mode page provides an application client with methods
to control the power condition of a
logical unit (see 5.10). These methods include:
a) Specifying that the logical unit transition to a power condition without
delay; and
b) Activating and setting of idle condition and standby condition timers to
specify that the logical unit wait for a period of inactivity before
transitioning to a specified power condition.
>>
Item a) implies to me that there is a method of telling the target to "go
to specified power condition immediately (i.e., now)".
Item b) implies to me that there is a method of specifying a timer value
for when to transition to the power condition.
Is the intent of listing a) separately to infer that a timer value of zero
is the "go immediately" command and that it is a one time event?  Or is the
intent that the timer value is modified to be zero and the side effect of
that modification is that the power condition is immediately transitioned
to?  Stated differently, does a mode select specifying a power condition
timer of zero modify that timer until the next mode select (or event that
changes mode parameters) thereby causing a device to transition out of and
back into that state on the receipt and completion of each command that
causes a transition (e.g., write)?
Thanks,
Kevin D. Butt
SCSI & Fibre Channel Architect, Tape Firmware
MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744
Tel: 520-799-2869 / 520-799-5280
Fax: 520-799-2723 (T/L:321)
Email address: kdbutt at us.ibm.com
http://www-03.ibm.com/servers/storage/
*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org



More information about the T10 mailing list