SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


    [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

    RE: iSCSI: SCSI Cmd PDU larger than 48 bytes


    • To: "Sanjay Goyal" <sanjay_goyal@ivivity.com>, <roweber@acm.org>
    • Subject: RE: iSCSI: SCSI Cmd PDU larger than 48 bytes
    • From: "Martin, Nick" <Nick.Martin@compaq.com>
    • Date: Wed, 30 Jan 2002 12:01:03 -0600
    • Cc: <ips@ece.cmu.edu>
    • content-class: urn:content-classes:message
    • Content-Type: multipart/alternative;boundary="----_=_NextPart_001_01C1A9B8.154FA02A"
    • Sender: owner-ips@ece.cmu.edu
    • Thread-Index: AcGpm/ho+pUWe1V0SOeT0AT39yfT3AADUrcA
    • Thread-Topic: iSCSI: SCSI Cmd PDU larger than 48 bytes

    Sanjay,
     
    This is a reasonable question.  Here is my opinion.
     
    Firstly the occurrence of such commands in most iSCSI sessions is not a random function with predictable distribution.
     
    For a target, the likelihood of a bi-directional command or a CDB larger than 16 bytes is near zero unless the target supports such commands.  Many of today's targets will support (understand and execute) neither bi-directional nor large CDB commands.
     
    For an initiator (HBA), the probability is zero until support is added to upper layer (device class) software to be able to send such commands.  Further the OS's interface to the HBA must comprehend and permit such requests to be passed from the device class drivers to the HBA drivers.   The HBA might not know whether device drivers will send such commands until they do so.
     
    Still these can not be effectively used unless the same customer has both.  For a bridge device which might be translating between iSCSI and Fibre Channel, again there must be support in both end points.  Otherwise the command will not be generated, or when generated and delivered, will be rejected.  Hopefully after the first such rejected command, additional similar commands will be very few and far between.
     
    There are SCSI commands defined today which require long CDBs or bi-directional data transfers and there are more commands being proposed which would require one or both.  At least one major vendor has already built devices which use them.  They may be widely used someday.  They are not yet "main-stream" or "common place".
     
    If a customer has a device which supports or depends on this functionality and an initiator which can use them, iSCSI must be able to deliver them.  In sessions with such a device, a high percentage of the commands could fall into these categories. 
     
    Thanks,
    Nick
    -----Original Message-----
    From: Sanjay Goyal [mailto:sanjay_goyal@ivivity.com]
    Sent: Wednesday, January 30, 2002 8:33 AM
    To: 'roweber@acm.org'
    Cc: ips@ece.cmu.edu
    Subject: iSCSI: SCSI Cmd PDU larger than 48 bytes

    what is the probability that the SCSI Cmd PDU will be more than 48 bytes? 

    the cases are:

    1. Bi-Di command

    2. CDB larger than 16 bytes

    Regards

    Sanjay G



Home

Last updated: Mon Jun 10 11:18:47 2002
10623 messages in chronological order