Understanding ACA

Kurt Chan kc at core.rose.hp.com
Wed Jan 4 09:34:46 PST 1995


From:  Kurt Chan, HP
To:    SCSI Reflector
Subj:  Understanding ACA
Date:  Wed Jan  4 09:08:28 PST 1995

In the Direct Disk Attach profile for FC-AL, we've attempted to
understand how ACA would be implemented.  While we don't mandate the
use or non-use of ACA, when it is used we want it to be interoperable.

It seems that there is a long history behind this feature, and that it
is possible that the disk drive, tape drive, and RAID communities may
have usages in mind which may not all interoperate with a host
attempting to implement a single ACA model.

For disk drives over FC, ACA would be useful to preserve Request Sense
data where the network does not provide guaranteed delivery of
autosense (e.g., FCP class 3).  In such cases, a specific limited
functionality for ACA is acceptable and perhaps even preferred in
order to simplify implementations.  Other device types, such as tapes,
may require fuller functionality during ACA conditions.

Two questions:

- Does X3T10 perceive a need to clarify the description of ACA for the
  purposes of interoperability?

- Is there a means for a Target to be required to retain sense data
  until it receives acknowledgement from the Initiator that the sense
  data has been delivered?  On a channel model, the REQ/ACK handshakes
  on the sense data may be sufficient.  However, on a FC network using
  FCP in class 3, the current definition of ACA is insufficient to
  ensure sense data preservation.  After several readings, it appears
  that there are still some channel-based assumptions in SAM and other
  X3T10 documents.

Weather permitting, I'll be in Lake Tahoe next week to discuss this if
any are interested.

Regards,
Kurt.




More information about the T10 mailing list