SFF-8449 / 8636 Conference Call Minutes, October 1, 2009

Olawsky, Barry Barry.Olawsky at hp.com
Mon Oct 5 11:48:13 PDT 2009


* From the T10 Reflector (t10 at t10.org), posted by:
* "Olawsky, Barry" <Barry.Olawsky at hp.com>
*
Minutes of SFF-8449 / 8636 conference call October 1, 2009
Attendance:
Mr. Min-Jie Chong			Agilent Technologies
Mr. Jim McGrath 		Cinch
Mr. Mickey Felton			EMC
Mr. Jim David			FCI
Ms. Katharine Schmidtke 	Finisar Corp.			  
Mr. Elwood Parsons		Foxconn Electronics		    
Mr. Barry Olawsky			Hewlett Packard Co.
Mr. Gourgen Oganessyan		Intersel
Mr. Harvey Newman			LSI Corp 
Mr. Gabriel Romero		LSI Corp.     
Mr. Brian Day			LSI Corp		      
Mr. Brad Besmer 		LSI Corp.		      
Mr. Galen Fromm 		Molex Inc.			    
Mr. Michael Rost			Molex Inc.
Mr. Chris Fore			NetApp	      
Mr. Alvin Cox			Seagate Technology	
Mr. Scott Shuey 		TycoElectronics 		    
Mr. Toney Chew			Volex, Inc.			  
Mr. Larry McMillan		WDC
Mr. Matt Davis			Zarlink Semiconductor		  
20 in attendance
Agenda:
Discuss proposed content for SFF-8636 (Common Management Interface) and
SFF-8449 (Mini Multilane Series Management Interface). SFF-8436 (QSFP+ Copper
and Optical Transceiver) is to be used as a basis for SFF-8636.
Discussion:
Barry Olawsky (HP) shared a shared SFF-8636 definitions and interface
signals. See bottom for attachment.
The first discussion topic was how to apply the interconnect terminology
"fixed" and "free" to the management interface. Barry Olawsky (HP) shared a
file which included "fixed" and "free" definitions from SFF-8644, a proposed
definition of fixed and free for SFF-8636 and graphical representations of
all interconnect components identified as fixed and free. Several suggestions
were made during the meeting that Barry will incorporate into his next draft.
Specifically,
1) Clearly indicate that the 2-wire management interface does not go from one
end of the cable to the other
2) Indicate, on the proposed diagrams, what segments of the interconnect are
within the scope of SFF-8636 and which are not (for example, 8636 will not
address high speed differential signaling or optical interfaces)
3) Provide a more detailed diagram (possibly a block diagram) of the
management interface circuitry that resides not only in the cable side but
also the system side. This could include the 2-wire slave and master
4) Replace "end-to-end" references to "free-to-free" or "fixed-to-fixed"
Continuing on with the definition section, it was recommended that the terms
"passive" and "active" change to "only requiring power for the management
interface" and "requires power in addition to what the management interface
circuitry consumes".
The discussion went off on a tangent to address what how wide designs (those
large than 4X) would be implemented. Molex proposed cost reducing the designs
by using only one paddle card and one non-volatile storage device. However,
the general consensus was that such a method would create additional
complexity in the memory map and the proposal was withdrawn for now.
Barry presented a list of all necessary SFF-8636 interface signals. Initially
his objective was to include only those signals common to all interfaces
referencing SFF-8636. That list included only the 2-wire clock and data would
be required for the management interface. However, Mike Rost pointed out some
signals of the signals perceived as optional would still have to be defined
since their functionality is documented in the memory map (for example,
interrupt). Barry will modify his proposal to account for this.
The meeting adjourned at ~11:10AM CDT
Call schedule:
Next conference call: October 8, 2009
Toll Free Dial in Number: (877)810-9442 
International Access/Caller Paid Dial In Number: (636)651-3190 
PARTICIPANT CODE: 3243413 
Webex information: 
https://seagate.webex.com/seagate 
Topic: SAS-2.1 PHY WG 
Date: Thursday, October 8, 2009 
Time: 10:00 am, Central Daylight Time (GMT -05:00, Chicago) 
Meeting Number: 826 515 680 
Meeting Password: newsas 
(Note: When logging onto WEBEX, please include your company following your
name. This will help the person recording attendance.)
----------    DISCUSSION MATERIAL    -----------
|FROM SFF-8644:
Fixed: Used to describe the gender of the mating side of the connector that
accepts its mate upon mating. This gender is frequently, but not always,
associated with the common terminology "receptacle". Other terms commonly
used are "female" and "socket connector". The term "fixed" is adopted from
EIA standard terminology as the gender that most commonly exists on the fixed
end of a connection, for example, on the board or bulkhead side. In this
specification "fixed" is specifically used to describe the mating side gender
illustrated in Figure 2.
Free: Used to describe the gender of the mating side of the connector that
penetrates its mate upon mating. This gender is frequently, but not always,
associated with the common terminology "plug". Other terms commonly used are
"male" and "pin connector". The term "free" is adopted from EIA standard
terminology as the gender that most commonly exists on the free end of a
connection, for example, on the cable side. In this specification "free" is
specifically used to describe the mating side gender illustrated in Figure 2.
PROPOSED SFF-8636 DEFINITIONS:
Fixed versus Free:
In this application the fixed side is one end of a point to point link
connected by one or more free side segments. Figure 1 illustrates how a
single free side segment connects two fixed side end points.
Figure 1
Figure 2 illustrates how multiple free side segments connect two fixed side
end points.
Figure 2
Passive:
A passive cable contains requires no power for an end to end connection to
operate properly. However, non-powered circuitry may be in the end to end
path.
Active:
An active cable requires power for the end-to-end path to operate properly.
*
* 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