Hi Dmitri,

This depends on so many factors that only you yourself can answer this. Your VM 
density all depends on the spec of your hardware, your overprovisioning of 
RAM/CPU and also very much on what size and workloads your VMs run and how much 
disk activity they have. When you on top of this make CloudStack management 
compete for resources with KVM on the same box then sooner or later one of your 
resources will start to run low – in your case it’s impossible to say without 
looking at the performance stats on your server – it could be CPU, RAM or disk 
I/O. A density of 450 VMs / server is not something we see often – I guess 
theoretically this could be achieved on light workloads and high spec servers – 
but most CloudStack users will run with lower figures than this.

Transactions from secondary storage to primary storage: in short the first time 
a template is used it is copied by the hypervisor from secondary (where it was 
uploaded to) to primary. The second time it is used CloudStack checks if the 
template is already in the chosen primary pool and if it is just uses the 
existing copy for linked clone. The Secondary Storage VM is mainly there to 
handle uploads and downloads to secondary storage as well as some snapshot 
operations.

Hope this helps,

Regards,
Dag Sonstebo
Cloud Architect
ShapeBlue

On 05/07/2018, 08:46, "Dmitry Berezhnoy" <d.berezh...@emzior.ru> wrote:

    Hello,
    
    I have installed cloudstack-management, cloudstack-agent(KVM hypervisor)
    and nfs-server at the same physical server. Then I have created virtual
    machines one by one in 8 parallel threads. For each thread I created shared
    network(VR). This server has two SSDs with ZFS filesystem. Copying files
    and disk speed is good enough (not only in system idle), but when I have
    450 virtual machines the next one has been creating during 5 minutes, is it
    correct work? This time becomes too long on 200 (about a minute) virtual
    machines and increase on every 10 machines or so. Does anybody have any
    ideas how to make it faster?
    
    Here is another question: how transaction from secondary storage to primary
    works. I see that images on primary are something like delta from original
    image (deduplicated?). Which component of Cloudstack is responsible for
    transfering data from secondary to primary?
    
    Thanks in advance for any ideas and info.
    
    Best regards,
    Dmitry
    


dag.sonst...@shapeblue.com 
www.shapeblue.com
53 Chandos Place, Covent Garden, London  WC2N 4HSUK
@shapeblue
  
 

Reply via email to