How do SAS self-configuring expanders self configure?

Elliott, Robert (Server Storage) Elliott at hp.com
Fri Jan 27 07:27:39 PST 2006


* From the T10 Reflector (t10 at t10.org), posted by:
* "Elliott, Robert (Server Storage)" <Elliott at hp.com>
*
Yes, it performs the full discover process.  If there is a 
configurable expander somewhere in the topology, the management 
application client in the self-configuring expander is as
responsible as any other for programming its routing tables.

If the domain contains only self-configuring expanders (how it
learns this is vendor-specific), then it has no interest in the
addresses found beyond a subtractive phy; it only needs to know 
the addresses below its table routing phys.

If we get rid of subtractive phys in SAS-2, then exploring the
whole domain will be necessary even in that case.

My feeling is that having all expanders perform just discovery 
(but never have to program anyone else's routing tables) yields
reasonable amounts of SMP traffic - one REPORT GENERAL per
expander + one DISCOVER per expander phy.  Better-optimized
functions like DISCOVER LIST will improve that.  Where
discovery takes an awful long time is when remote
application clients are configuring routing tables, adding
one CONFIGURE PHY ROUTE INFORMATION per table entry (e.g.
hundreds/thousands of SMP function calls).  Again, a LIST 
function could help optimize that, but it's still more 
traffic than not configuring at all.

--
Rob Elliott, elliott at hp.com
Hewlett-Packard Industry Standard Server Storage Advanced Technology
https://ecardfile.com/id/RobElliott


 

> -----Original Message-----
> From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf 
> Of Ralph Weber
> Sent: Thursday, January 26, 2006 11:11 PM
> To: t10 at t10.org
> Subject: Re: How do SAS self-configuring expanders self configure?
> 
> * From the T10 Reflector (t10 at t10.org), posted by:
> * Ralph Weber <roweber at ieee.org>
> *
> Rob,
> 
> I figured you would know where to find the 4x4. Congratulations.
> 
> "... and shall contain a management application client and
> SMP initiator port to perform the discover process to configure
> its own expander route table."
> 
> What discover process? Surely not the "... level-order (i.e.,
> breadth-first) traversal of the SAS domain" specified in 4.7.1.
> That would seem to be a bit over the top, particularly in the
> case of an expander near the bottom of the tree.
> 
> Have the performance implications of every expander in a large
> configuration doing this been reviewed lately?
> 
> All the best,
> 
> .Ralph
> 
> Elliott, Robert (Server Storage) wrote:
> 
> >* From the T10 Reflector (t10 at t10.org), posted by:
> >* "Elliott, Robert (Server Storage)" <Elliott at hp.com>
> >*
> >A self-configuring expander is one that fills in its routing 
> tables on
> >its own, and does not expose them through REPORT/CONFIGURE ROUTING
> >INFORMATION functions.
> >
> >4.1.5 describes the two types:
> >"Expander devices with expander phys with the table routing attribute
> >contain an expander route table. The expander route table may be
> >configurable. An expander device with a configurable route table
> >depends on a management application client within the SAS domain 
> >to use the discover process (see 4.7) to configure the expander 
> >route table.
> >
> >An expander device with expander phys with the table routing
> >attribute that does not have a configurable route table shall be 
> >self-configuring, and shall contain a management application client
> >and SMP initiator port to perform the discover process to configure 
> >its own expander route table."
> >
> >What more does the standard need to say?  
> >
> >The details of the route table implementation are vendor-specific.
> >The addresses it must add (in whatever manner) are defined by 
> >the discover process.
> >
> >One detail lacking is a way for self-configuring expanders to
> >report their status with regard to error handling.  We agreed 
> >to deal with that in SAS-2.
> >
> >We briefly discussed adding a generation count to all the
> >SMP functions so a management app client would realize that
> >configuration started changing since it last checked the
> >REPORT GENERAL function CONFIGURING bit, but decided the
> >BROADCAST (CHANGE) when configuration stops was adequate
> >(it retriggers a new discover process by that app client,
> >which will pick up a consistent new set of results).
> >
> >More than one vendor has implemented self-configuration in
> >a real expander, and I haven't heard any complaints about not
> >understanding what to do.
> >
> >--
> >Rob Elliott, elliott at hp.com
> >Hewlett-Packard Industry Standard Server Storage Advanced Technology
> >https://ecardfile.com/id/RobElliott
> >
> >
> > 
> >
> >  
> >
> >>-----Original Message-----
> >>From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf 
> >>Of Ralph Weber
> >>Sent: Thursday, January 26, 2006 4:10 PM
> >>To: T10, Reflector
> >>Subject: How do SAS self-configuring expanders self configure?
> >>
> >>* From the T10 Reflector (t10 at t10.org), posted by:
> >>* Ralph Weber <roweber at ieee.org>
> >>*
> >>I have been told several times that SAS-1.1 (and by inference SAS-2)
> >>define how self configuring expanders do their self configuration
> >>thing. I have taken a stab at finding this definition:
> >>
> >>  http://www.t10.org/ftp/t10/document.06/06-087r0.pdf
> >>
> >>The only 'self-configuring expander devices shall ...' I can find
> >>in SAS-2 (give or take a few words) is:
> >>
> >>  Self-configuring expander devices shall originate a BROADCAST
> >>  (CHANGE) when they are done configuring.
> >>
> >>Anyone with greater insight than I, is welcome to engrave the SAS-2
> >>clause number for the location of these mythical self configuration
> >>requirements on a 4x4 and beat me about the head and 
> >>shoulders with it.
> >>
> >>On the other hand, the next person who claims that SAS-x defines how
> >>self configuring expanders work is advised to arrive with such a 4x4
> >>in their hip pocket in order to have any hope of being taken 
> >>seriously.
> >>
> >>All the best,
> >>
> >>.Ralph
> >>
> >>
> >>*
> >>* For T10 Reflector information, send a message with
> >>* 'info t10' (no quotes) in the message body to majordomo at t10.org
> >>
> >>    
> >>
> >*
> >* For T10 Reflector information, send a message with
> >* 'info t10' (no quotes) in the message body to majordomo at t10.org
> >
> >
> >
> >  
> >
> 
> *
> * For T10 Reflector information, send a message with
> * 'info t10' (no quotes) in the message body to majordomo at t10.org
> 
*
* 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