Is it possible that VM guest using more resource RAM than its allocation.?
I am running 2 VMs. Each VM has no fancy setting. It just straight forward setting via web GUI.
This node has 64GB of RAM
Specification for node and VM guest are below
Node :
EMBY Guess (Windows 10) :
CPU 8 cores, RAM 31
Veeam Backup Guess (Windows 2019 server)
CPU 6 Cores, RAM 16
Problem was when running those 2 VMs together. I/O Delay was about 20%-30%. EMBY was fine but VEEAM was real slow. I meant super super slow when doing backup job. Some of the job failed.
The problem solved when I shut down EMBY server. VEEAM Backup just working great
I think it might come from 3 possibility.
First, VM guess or node itself somehow using more resource than its allocated (maybe in some condition which I don't know).
Second, I/O issue. I am not sure about this because EMBY using hdd internally. I meant in the node itself. However, VEEAM backup to another hdd via NFS. Basically, They are using a difference HDD. EMBY using SAS 15K RPM enterprise grade. NFS using HDD consumer grade.
Third, I over commited CPU core. E5-2430 has 6 core 12 thread for each. It shows in the node that has 24 cores. I assign the total of 12 cores for both VMs.
I am not sure how the core count in this case. 14/10 or 14/24
I want to make sure the cause of this before I buy new SDD.
Any idea? Thank you for sharing.
I am running 2 VMs. Each VM has no fancy setting. It just straight forward setting via web GUI.
This node has 64GB of RAM
Specification for node and VM guest are below
Node :
CPU usage 10.48% of 24 CPU(s) | IO delay 0.04% |
Load average 2.51,2.00,1.13 | |
RAM usage 16.08% (15.16 GiB of 64.33 GiB) | KSM sharing 0 B |
/ HD space 0.22% (2.16 GiB of 976.64 GiB) | SWAP usage N/A |
CPU(s) 24 x Intel(R) Xeon(R) CPU E5-2430 0 @ 2.20GHz (2 Sockets) | |
EMBY Guess (Windows 10) :
CPU 8 cores, RAM 31
Veeam Backup Guess (Windows 2019 server)
CPU 6 Cores, RAM 16
Problem was when running those 2 VMs together. I/O Delay was about 20%-30%. EMBY was fine but VEEAM was real slow. I meant super super slow when doing backup job. Some of the job failed.
The problem solved when I shut down EMBY server. VEEAM Backup just working great
I think it might come from 3 possibility.
First, VM guess or node itself somehow using more resource than its allocated (maybe in some condition which I don't know).
Second, I/O issue. I am not sure about this because EMBY using hdd internally. I meant in the node itself. However, VEEAM backup to another hdd via NFS. Basically, They are using a difference HDD. EMBY using SAS 15K RPM enterprise grade. NFS using HDD consumer grade.
Third, I over commited CPU core. E5-2430 has 6 core 12 thread for each. It shows in the node that has 24 cores. I assign the total of 12 cores for both VMs.
I am not sure how the core count in this case. 14/10 or 14/24
I want to make sure the cause of this before I buy new SDD.
Any idea? Thank you for sharing.