Re: Towards Consensus on TCP connections

[julian_satran@il.ibm.com]

Jeff Fellin wrote...

------------------
If we don't use the SCSI-GPP than I propose changing the one LUN per
TCP connection group to be a SCSI target id that multiplexes all LUN's
of the target id onto the connection group. I believe this will decrease
the amount of connections and provide data channels for large transfers.
_______________

Thanks Jeff, We went through the same documents and reached the same
conclusion.

Errors are handled better that in the basic parallel SCSI and it has a neat
CRC mechanism
at predefined points in the stream. We hope to achieve the same result with
a digest+CRC
mechanism.

The sentence quoted above from your note ... I am not sure that I
understand it.
The LUN in iSCSI is carried in the packet and the target-id is mostly
implied except for
third party commands (and the name sent with the login).

What did you have in mind?

Regards,
Julo