> I know there are better solutions like RAID controllers with their own 
> microcontrollers,

the microcontroller is usually orders of magnitude slower than the host,
so "better" does not apply.  offloading parity makes a certain amount 
of sense, but it becomes nontrivial where you draw the line (if you have 
a parity engine, it's nice to have a substantial cache.  if you have a 
big cache, you have to add more smarts in managing it -> TCQ -> pretty
soon you have a full blown SAN raid controller (and the mediocre performance
that implies.)  I think it's mainly marketing pressure that drives people
away from the wisdom of simple-is-better.

> a solution which does not consist of a software-raid managed by the 
> kernel (I had some trouble with it when one of the two disks (RAID 1) 
> crashed - all data lost).

don't be silly - md raid is used in probably tens of millions of servers 
and is not flakey.

-
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to