Using a mixture of k-sizes to fill up a drive seems to be a bad idea

General Chia Discussion
1

“Take some more tea,” the March Hare said to Alice, very earnestly.
“I’ve had nothing yet,” Alice replied in an offended tone: “so I ca’n’t take more.”
“You mean you ca’n’t take less*,” said the Hatter: “It’s very easy to take* more than nothing.”

There’s always a lot of chatter about k-sizes. The choice of the size is very prominent, and I have seen a lot of places trying to explain what it’s for. So far the most reasonable explanation is that you can use it to fill up your drives completely by picking just the right mix of plots.

There are tools out ther for figuring out the right mixture to fill drives completely, and even the official faq says this:

There is only one reason why you might want to plot larger than k=32 and that would be to maximize the total utilization of a given drive or space. A couple of k=33 plots with a majority of k=32 plots can reduce the amount of leftover unused space on a drive.

at the same time, the Windows Tips section says this:

Leave free space on your drive

If you end up having your HDD or SSD have any issues that require disk repair, you will need to have free space on the disk that is larger than the largest file. For k32 plots, you will need to leave > 101 GB. This way, if your plots ever have errors (as reported by plot check tool), you will at least be able to try to repair them with CHKDSK /r. However, CHKDSK cannot repair files that are larger than the remaining free space on a drive, and will have an error

I don’t know if fsck can repair errors on a drive with less than a file’s size space left, but it seems like filling up a drive completely is a bad idea, especially filling it up completely with different k-sizes.

2 Likes
2

I tend to ignore stuff like this (getting juuust the right number of files to exactly fill a disk) under the banner of “you’ll spend so much time micro-optimizing that you’ll fail to see the big picture”.

Micro-optimizations tend to be of limited benefit, and occupy a disproportionate amount of your brain… in my experience, anyway :wink:

4 Likes
3

I’ll have to check but I’m pretty sure ext4 doesn’t have this problem, pretty sure I’ve fscked disks with only 8GB free. Sounds like a mess.

Also you probably don’t need to fsck much after you have filled the drive up. Those bits and inodes are never changing again… If you have errors it’s probable the whole drive needs to be replaced anyway.

Edit: also mechanical hard drives can get slow in normal use cases as they approach capacity due to files getting increasingly fragmented and being forced to use the slow end of the platter.

But with plot files and -T largefile4, this is absolutely not a normal use case. There’s no reason to worry about that kind of thing.

1 Like