SSD units are fast, well, depending upon design, controller and other things. Sandforce units use a compression and overprovision technology to reduce write amplification. SSD units do writes, optimally, in erase block sizes.
This suggests that your RAID chunk size should be a multiple of the erase block size. This is a good thing.
The issue is that if you have a hardware RAID controller, you might think that the optimal way to handle this is to build a RAID5 or RAID6 atop this SSD pool. Then aggregate the RAIDs into a RAID50 or RAID60.
The RAID chips aren’t able to keep up with the SSDs though. We can pretty much max out each RAID core bandwidth with 5-6 modern SSDs. And RAID5 or RAID6 isn’t terribly good for high IOP circumstances.
I am looking at some more exotic layered RAIDs, which change the dynamics a bit.
Will see if my experiments net anything interesting tomorrow.
Viewed 25570 times by 4569 viewers