Hacker News Reader

Comment by fps_doug | Hacker News Reader

I felt a bit stupid; I re-read the important parts several times but failed to visualize what that looks like on the disks, and how blocks are reconstructed on disk failure. A little googling lead me to a paper by Intel[0] which has a simple visualization of how this works on page 11, in case I'm not the only one.

[0] https://docplayer.net/117362530-Draid-declustered-raid-for-z...

View on HN

Yeah that is a good link; I especially liked the text on page 7.

I believe Windows Storage Spaces has long used a similar technique[0] too, though it has been a few years since I was looking into it. I wonder if ZFS DRAID will eventually allow full utilization of disks of mixed sizes too, that was always a very attractive feature of Storage Spaces to me.

[0] https://techcommunity.microsoft.com/t5/storage-at-microsoft/...

Oh? When I first saw that illustration it hugely frustrated me about how poor that illustration is.

The numbers are not explained, nor are the rows, or columns. And the numbers repeat.

I saw the diagrams first outside of that pdf, though. Still, the way it goes "4, 8, 2, 16, 1, and so on..." is a boatload of "wat?".

I think this quote makes it clearer: "unlike the raidz1 configuration shown above, in the dRAID1 configuration the neighbor drives of the failed drive 0 (i.e. drives in a same data+parity group) are not fixed".

So the RAIDZ is set up with fixed groups of drives, but the dRAID is not, even though the parity ratio is the same.

Yeah, I get it. But the illustrations I found terrible. Those images did not speak a 1000 words. In fact they are just confusing.

Your comment explains it better than the illustration.

Ah, so a distributed redundancy drive instead of a dedicated one as with normal raid.

Reply on news.ycombinator.com