On Wed, Feb 8, 2012 at 2:56 PM, Kurt Buff <[email protected]> wrote:
> There may be a workaround, by chaining together some LUNs (perhaps in
> Windows, too), but that seems like a fugly hack to me.

  Hmmm, that's interesting.  If the problem is that their iSCSI
implementation is written for the SCSI-2 command set, then aggregating
LUNs *in the device* wouldn't help.  You'd still be stuck with a
32-bit LBA, as far as I know.  (Appending LUNs in Windows would be a
different story.)

  As far as "ugly hack"... if this was a legitimate limitation, I'd
say aggregating LUNs would be an appropriate solution.  If a single
disk isn't big enough, we aggregate multiple disks (and call it
"RAID").  Using the same strategy for a limitation at the SCSI
protocol level that does not exist at a higher level of the OS is a
very legitimate approach.

  That said, I suspect this is a purely artificial limitation.  As
noted, SCSI-2 is twenty years old.  I believe SBC-1 introduced 64-bit
LBAs, and that was released in 1997 -- fifteen years ago.  The iSCSI
specification appears to be RFC-3720, which wasn't published until
2004.  There's no way EMC was unaware of this.

>>> it's truly a bad decision to hold back on it for market differentiation.
>>
>>  Your company bought it anyway, so maybe not.
>
> And I'm bitching about it on a list that is widely read, and where I
> have a modicum of respect afforded my opinions, so I'll hold fast to
> my thought on that...

  We can hope.  In my experience, PHB trumps BOFH every time, but we
can hope.  "Even the losers get lucky sometimes."[1]

-- Ben

[1] This statement is intended to be inclusive of the author.

~ Finally, powerful endpoint security that ISN'T a resource hog! ~
~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/>  ~

---
To manage subscriptions click here: 
http://lyris.sunbelt-software.com/read/my_forums/
or send an email to [email protected]
with the body: unsubscribe ntsysadmin

Reply via email to