Device Sharing
Gary A. Hoffman
gary.hoffman at skipstone.com
Wed Feb 26 07:38:06 PST 1997
* From the SCSI Reflector (scsi at symbios.com), posted by:
* "Gary A. Hoffman" <gary.hoffman at skipstone.com>
*
g'mornin Dave,
Something you might want to look at here also ..
SBP-2 has defined (what might be a different way)
a methodology for device ownership.
g
At 08:29 AM 2/26/97 CST, Dave Peterson wrote:
>
>*
>* From the serial_solutions Reflector, posted by:
>* dap at anubis.network.com (Dave Peterson)
>*
>* To post to this Reflector, send email addressed to
serial_solutions at zk3.dec.com
>*
>
>Hello All,
>
>The purpose of this message is to generate discussion about the sharing
>of devices (specifically tape transport's) between multiple host's and to
>convey the need for consistency between the various standard bodies
>regarding the usage of reserve and release.
>
>The sharing of transport's is becoming more desirable as companies migrate
>"open system" machines to the traditional glass house for data and storage.
>In the open system arena there is no consistent or well-defined method
>to share a tape device. Tape devices have reserve/release commands to
>facilitate sharing, but few (if any) device drivers
>issue a reserve at open and a release at close. Basically it is left to
>the application.
>
>The SCSI-2 spec (at least the one I have) states that the reserve and release
>commands are mandatory for sequential devices. After looking at SSC rev 9 I
>noticed the reserve and release commands are now optional for
>sequential-access devices. Currently PLDA_18 requires reserve and release
>for sequential-access devices.
>
>I believe reserve and release should be mandatory for sequential-access
>devices and device driver's should issue a reserve at open and release at
close.
>The ability to detect the device is already reserved at open and not release
>the device at close time (int the device driver) would also be beneficial
>for applications that open and close the device multiple times during it's
>processing.
>
>With the advent of Fibre Channel and SCSI-3 I believe now is a good time to
>address these issues.
>
>===================================================================
>Dave Peterson phone : 612-391-1008
>Senior Engineer
>Network Systems Corp. email: dap at network.com
>
gary a hoffman
Skipstone, CEO
512.305.0200 main Skipstone Is 1394
512.305.0212 fax
gary.hoffman at skipstone.com 3925 West Braker Lane, Suite 425
http://www.skipstone.com Austin, Texas 78759 USA
*
* For SCSI Reflector information, send a message with
* 'info scsi' (no quotes) in the message body to majordomo at symbios.com
More information about the T10
mailing list