Since I have to assume that NetApp would use a high performance
architecture
for the benchmark I am guessing that this is the way to go, but there is
quite
a bit of detail missing.
Is it a good or bad idea to split a raid group over two FC-AL
interfaces?
Somewhat bad... as has been talked about here before, there is a small write penalty introduced, on the order of 10%, which only introduces itself noticeably under heavy write loads. However, you should avoid it if you can.
What does "striped across both disk controllers" mean?
It means the filesystem was striped across both disk controllers. I does not say any raid group was striped across disk controllers. I suspect that they had all one filesystem, two raid groups, one raid group per each controller.
If there are spare disks (same size) on both FC-AL interfaces and a disk
fails
Which one is used to rebuild? (Same FC-AL or random choice)
That I don't know. I think someone brought this up here before.
Is it a good or bad idea to split a volume over multiple FC-AL
interfaces?
Neither. I guess, in some sense, it's slightly bad, since if you lose a particular controller you would only lose volumes on that controller, but my guess is in such event you are looking at a reboot or a failover situation anyway.
Bruce