Going OT For David - RE: [Mimedefang] mail server performencedeclain

[David F. Skoll]

Ben Poliakoff wrote:

> Let's also remember that the IO bandwidth available to this device is
> limited at the top end by the SATA specification.  In other words while
> this device is going to be faster than spinning platters it's not going
> to be nearly as fast as a true "ram disk".

Yes, but you have essentially 0ms seek time.  It's typically seek time,
not I/O bandwidth, that kills disk performance on a mail server.

Regards,

David.