Power-Loss-protected cache size on SSD: Yes, finally someone reveals the size (Kingston)

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

MrCalvin

IT consultant, Denmark
Aug 22, 2016
87
15
8
51
Denmark
www.wit.dk
Beside the obvious reason of PLP it should also help synchronous I/O performance e.g. SQL (and all ESX guest writes I believe!), as when data are written to the quick PLP protected disk-cache the disk rightfully can report back that the data IS written to the disk and the system can move on.
But then the cache size obvious play a big role! And Seagate, Micron and Kingston don't give that vital information on their SSD SATA disks with PLP, and all the storage reviewers out there unfortunately don't have focus on this and don't demand to get this information from the vendors :-(

I've asked Seagate, Micron and Kingston but the two former rejected to tell the cache size, but Kingston did!

I expected it was because the cache-size was very small and that the whole PLP feature of these rather cheap "data-center" drives was just marketing scam, and I actually still think so in regards of Seagate and Micron (their support are extremely incompetent, it's an embarrassment to witness).
5300pro.jpg ironwolf_110.jpg
Micron 5300 Pro Seagate IronWolf 110

But yesterday I actually got a good answer form Kingston, though they had to dig a little into the organization to get the information.
ktc-product-ssd-dc500m-sedc500m480g-2-lg.jpg
But here it is:
SEDC500R 480GB – 512MB
SEDC500R 960GB – 1GB
SEDC500R 1920GB – 2GB (cheapest price I could find: €340 ex. vat)

SEDC500M 480GB - 1GB (cheapest price I could find: €113 ex. vat)
SEDC500M 960GB - 2GB
SEDC500M 1920GB - 4GB (cheapest price I could find: €450 ex. vat)

That's pretty decent sizes, way more than expected, and I suppose they can actually make a difference. Writing 1GB of SQL data that's pretty much.
(if the numbers are true)

My respect to Kingston just increased significantly.
I did reply them I was positively surprised and that there were no reasons to be ashamed of those number :p, and that they should update their datasheets (not that I think it will happen!)


I don't have any benchmark to back it up, don't have any of the drives yet, so who knows how it will play out in the real world...maybe any of you have any thoughts on the matter :)