FASTER SCSI
Gene Milligan
Gene_Milligan at notes.seagate.com
Fri May 26 06:59:29 PDT 1995
Jim wrote:
>One way is to change arbitration to decouple the number of
>devices from the number of data lines. This is straightforward,
>but is obviously a backward compatibility problem (all devices
>must know the new protocol (actually there may be some
>trick to support some legacy devices)).
Before playing tricks, take a look at SCAM. In the PlugnPlay SCAM environment
your allowed to have physical address conflicts. I have not looked but there
might be a way to enhance SCAM to assign addresses in a manner that uses more
addresses than bus lines. The scheme might be to assume a maximum of eight
legacy (broader than SCAM legacy) devices and require that they be assigned
addresses 0-7. Then the funny arbitration would be done with the bits 8-15
allowing up to 256 non legacy devices plus the 8 legacy devices. Obviously we
would run out of electrical compatibility long before we reached this protocol
limit.
I did not mean to imply that I was actually going to look into this embryonic
idea. I merely let the bears chase me back into the cabin.
Gene
More information about the T10
mailing list