4th phase - DIsk performance

Hello folks,

Is this normal?

The disk was still operating at 100% at 15:21 for the 8 plots i set previously.

I also attached the 8 plots from the file. (14:38->14:41)

what i wanna ask is , when the plots appeared in the HDD final destination at (14:38->14:41) , does it mean it is already fully plotted?

It is because Chia Hanged, when i came back to my computer on 15:21 only to check the disks operating at 100% still.

If you’re trying to write 8 plots at the same to an external USB drive, that is going to bottleneck severely. I’m not even sure when it will finish, could be hours on hours.

1 Like

A k32 on my setup takes ~16m to complete phase4 in solo and from external drive SSD to external drive HDD.
The k33 takes ~22m. Interesting fact that k33 is double size of a k32 but it does not take double the time.
On your case, copying 8 plots at same time to external HDD it will clog the USB pipe. You will need to adjust the delay timings between plots.
That is why is important to start simple/small and understand how the system behaves. Then add small incremental improvements to see the difference.

Hope it finishes safely and your plots are good.
NOTE: don’t forget to double check the plots with chia plots check command.

Cheers!

Or could i do ssd1 (2tb → hdd1) 4 plots

Ssd2 (2tb → hdd2) 4 plots

So far parrellel plotting of 8 (4x2) together to 1 hdd → takes around 8-10 hours…
i wasnt sure on official timings as i usually leave it n go out or go and sleep and when i wake up, it is usually done.

If it was 8 hours for 8 plots, 1 day 24hours i can do around 24 plots.

Is that align to my 8gb 16 cores setup

You can try measure the timings by just copying files that way. Surely splitting things is better than sending everything to same HDD.
Be aware that when you’re copying plots the farming HDD, your harvester will start complaining with longer times to check for profs.
Check the logs if you’re being it by the 30s threshold.

1 Like

I have 8 external hhd. Shld I run 4 plots with ssd1 → segregate it to 4 different hdd.

4 plots ssd2 → segregate to 4 diffrenet hdd.

Or just 2 hdd for now…

All of them are 16 tb externals

If you are running them at the same time without delay, then segregation is the way to go. I would start with 2 HDD for now.
If you are parallelizing with delay, check the plotting logs to adjust the delay so that each phase4 copy process does not overlap.
Hope this makes sense.

Noted. Thank you for the advise. How do we know if the plots that are farming are good? Do i click on them to see the log? I dont have access to edit the cli yet. Having problems editing. So far, been just using the Gui. Are are farming plots by default good or working properly?

You can only do it on the CLI AFAIK.
.\chia.exe plots check (windows)
chia plots check linux/macOS

Here is the documentation:

1 Like

i started cli.

what are useful commands, if buckets are running line by line = plotting?

if so, how can i use the cli when it is plotting

I have started to use cli.

When i plot with cli 4 plots, it doesnt appear in the gui. Im also plotting from gui. Will it conflict?

N regarding farming plots. Currently my hdd1 is where all the plots are. I have added hdd2.

My current plotting is 4 to hdd1 n 4 to hdd2. Since u mentioned that this might affect farming as the farmining hdd folder will be slowed. Shld i have hdd1 for all the farming plots while hdd2 n hdd3 get 4 plots each? And
Please advise.

So for my scenario where i dont use raid 0 to combine my 2 Ssd’s. I have to open 2 cli commands to run 4 plots each right?

Also i have to open multiple cli if example i wanna check plot checks

Or any other commands right?

Please enlighten me