Re: iSCSI: Session Partial Resolution

[Michael Krause <krause@cup.hp.com>]

At 02:05 AM 9/20/00 -0700, Matt Wakeley wrote:
>Black_David@emc.com wrote:
>\
>The "iSCSI Requirements" document requires this capability:
>
>"[R] High bandwidth, bandwidth aggregation.
>
>[D] The bandwidth (transfer rate, MB/sec) supported by storage
>controllers is rapidly increasing, due to several factors: (1)
>Increase in disk spindle and controller performance; (2) Use
>of ever-larger caches, and improved caching algorithms; (3)
>Increased scale of storage controllers (number of supported
>spindles, speed of interconnects). Not only must the iSCSI
>provide for full utilization of available link bandwidth, it
>also must exploit parallelism (multiple connections) at the
>device interfaces and within the interconnect fabric."
>
>
>Unless there is "consensus" that this is not required, I think the multiple
>connections/session must remain. If nothing else, at least the "hooks" should
>remain in the document (such as for login, etc) so as to minimize
>incompatibilities with future versions of the specification.  The mechanisms
>to "bundle" tcp connections into a session are the same regardless of whether
>they're used "synchronously" or "asynchronously".  Also, it's much easier to
>ignore unimplemented fields in a header than to add fields after the fact.

I'd rather pre-enable for the future which does not really impact today's 
implementation costs and avoid a major rework later, hence I concur with Matt.

Mike