AMD CPU can't plot in parallel?

[ganezasan]

I'm using Ryzen 3900x and try to make 11 plots, but the CPU doesn't use full resources and CPU usages are always around 20%. Should I move to Intel CPU? or Is there any way to use full CPU resources?

• CPU: Ryzen 3900x (12 cores)
• SSD: 3TB RAID0
• Memory: 80GB

Here is my script.

#!/bin/bash
screen -d -m -S chia1 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 0h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp1  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia1_1_.log'
screen -d -m -S chia2 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 1h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp2  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia2_1_.log'
screen -d -m -S chia3 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 2h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp3  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia3_1_.log'
screen -d -m -S chia4 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 3h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp4  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia4_1_.log'
screen -d -m -S chia5 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 4h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp5  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia5_1_.log'
screen -d -m -S chia6 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 5h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp6  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia6_1_.log'
screen -d -m -S chia7 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 6h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp7  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia7_1_.log'
screen -d -m -S chia8 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 7h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp8  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia8_1_.log'
screen -d -m -S chia9 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 8h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp9  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia9_1_.log'
screen -d -m -S chia10 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 9h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp10  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia10_1_.log'
screen -d -m -S chia11 bash -c 'cd /home/user/chia-blockchain && . ./activate && sleep 10h && chia plots create -k 32 -b 4000 -e -r 4 -u 128 -t /mnt/ssd-raid/temp11  -d /mnt/disk-hdd-1 | tee /home/user/chialogs/chia11_1_.log'

[cccat6]

It's not related to what CPU you are using.
You should check if all the tasks are running and check your shell.

[ganezasan]

@cccat6 Thank you for your message!
It seems disk I/O might cause a problem, Raid 0 constructed by 4 SSD, and it seems one SSD is a bit slower than others and it makes all the plotting be slower.

$ sar -p -d 1 1
Linux 5.4.0-72-generic (taka-chia-plot) 	05/09/21 	_x86_64_	(24 CPU)
...

Average:          DEV       tps     rkB/s     wkB/s     dkB/s   areq-sz    aqu-sz     await     %util
Average:          sda     91.00   6676.00   1184.00      0.00     86.37      5.87     66.48     92.00
Average:          sdb     96.00   6656.00   1089.00      0.00     80.68      0.00      0.52     17.20
Average:          md0    759.00  31476.00   8433.00      0.00     52.58      0.00      0.00      0.00
Average:          sde     88.00   6676.00   1100.00      0.00     88.36      0.03      0.90     18.80
Average:          sdd     91.00   6616.00   1136.00      0.00     85.19      0.00      0.48     16.80

I removed a slow SDD from Raid0 and it's been good so far.

↓

Average:          DEV       tps     rkB/s     wkB/s     dkB/s   areq-sz    aqu-sz     await     %util
Average:          sdb    230.00  25600.00  68064.00      0.00    407.23      3.21     15.79     89.60
Average:          sde    241.00  23040.00  67060.00      0.00    373.86      1.86      9.39     76.40
Average:          sdd    239.00  23552.00  65032.00      0.00    370.64      1.91      9.79     70.40
Average:          md0    931.00  74240.00 345976.00      0.00    451.36      0.00      0.00      0.00

[cccat6]

Raid0 won't be affected by slower drive. Make sure it is raid0 but not other raid types. Also, don't use software raid, it will eat your CPU and limited by single-thread performance. Unless you have redundant CPU resources and raid does not use up them. It should be some other factors that affected your plotting.

[ganezasan]

It looks like changing Raid0 doesn't fully solve this problem. Process status often shows D uninterruptible sleep (usually IO), and they can't use much CPU resources.

This [plotting guide] recommends using mdadm to configure RAID0, but do you think RAID should configure by BIOS function? I don't know this is an official guide but I found this on Chia docs page.

https://docs.google.com/document/d/18a5_MO88hv_DL_644OGFoCiKnpR9udl-PAWKlyIqqQU/edit#

Do you have any ideas which I should try?

[ganezasan]

I configured AMD RAID but it doesn't solve this problem.
I tried Intel CPU and it much faster with the same OS and same configuration.

[BrandtH22]

If you are still experiencing the issues it is recommended to upgrade to the latest version of chia. Please download it from here: https://www.chia.net/downloads/

Since this issue has been open for a number of years without additional comments, we will be closing this ticket but if we have closed this ticket in error do not hesitate to reach out to us again with any followup questions, comments, or if the issue persists after an update.

Please note that upgrading major versions of chia generally requires the config.yaml to be recreated using these steps:
1. Stop chia
2. Rename the config or move it to a different directory
3. In CLI run chia init
4. Copy any custom settings from the old config to the new (plot drive directories, pool info, trusted node info, etc) - be careful here as yaml is very picky about formatting including those leading spaces
5. Start Chia

The best place to reach our support team is on Discord (https://discord.gg/chia) or by reopening this ticket.