RAID Results and M.2 Types
On the left, we have one KC1000 for comparison, and on the right, we have our performance on our RAID 0 array. We aren't able to double our sequential speeds because of the DMI's 3.5GB/s limit, but we do see noticeably higher sequential speeds. We also see that our 4K reads/write didn't take a hit but remained roughly the same, which is typical of a RAID0 array.
Going a bit off topic, not all of SuperMicro's (and other vendors) motherboards have their M.2 slots directly routed to the PCH like on the motherboard we used. Many motherboards (most) have M.2 slots that share bandwidth with other devices (mostly SATA or an x4 PCI-E slot). We wanted to see if the different types of M.2 drive connections interfered with drive performance. Quick switches can cause performance to drop a tiny bit, but let's see how much, just for fun!
We took another brand's motherboard and ran it through the same tests using Anvil Storage benchmark, which is a bit tougher on the drives than CrystalDiskMark. The 'Two Shared + One Linked' should provide the best results, and the 'All Shared + Two Shared', which goes through the most switches, should produce worse results. We see that sequential and 4K writes actually seem to take a small hit, while reads don't really change. Overall though, we found that the types of M.2 slots aren't super important, but if you know the path your M.2 slots data travels, it's best to take the most direct route to the PCH.
Last updated: Nov 15, 2019 at 12:11 pm CST
PRICING: You can find products similar to this one for sale below.
United States: Find other tech and computer products like this over at Amazon.com
United Kingdom: Find other tech and computer products like this over at Amazon.co.uk
Australia: Find other tech and computer products like this over at Amazon.com.au
Canada: Find other tech and computer products like this over at Amazon.ca
Deutschland: Finde andere Technik- und Computerprodukte wie dieses auf Amazon.de