Just my blog

Blog about everything, mostly about tech stuff I made. Here is the list of stuff I'm using at my blog. Feel free to ask me about implementations.

Soft I recommend
Py lib I recommend

I'm using these libraries so you can ask me about them.

The conclusion after migration VMvsHyperV

So, what can I say before migration? The first day looks good, nothing leaks, all services work as expected, and have some additional configuration for MAC addresses etc. During migration, I accidentally lost one system disk for one of my VM and photos backup disks, but this is not an issue, because all important stuff has more than one backup.



  • The first expression about Hyper-V - it's more thrifty to system resources, I'm still tuning it to make my vms run smoothly but this is the first thing I saw.
  • Then, Hyper-V has a good mgmt console for all my needs, I have a small env. so I do not need a HUGE MGMT-VM to host vCenterServer because of the new VM ver. cannot be managed just from preinstalled vC-client.
  • Linux Vms such as Centos 6, Debian etc look good and running happily, the only change I've made - is bringing up eth0 again, because it was down (network hardware changes)
  • The BIG plus to hyper-v for small env - can use all resources till vms are powered off! This is very important to make backups and convert virtual disks, file systems, etc.

It was so useful when I had a different powerful tower PC to run all my VM conversions - from vmdk to vhd, and run disk defrag. etc. And if you have one hardware server for all your needs this is exactly all you need, you can have a powerful, running machine, which can use all RAM and CPU for all your migration, backup or other needs. When I use ESXi, I can't run mgmt tasks till all my vms are powered down, also I must keep my vCenterSever on another PC and keep it connected to the ESXi host in any way.

  • Can mount all VHD - this can be useful to manage broken VMs or keep data if the virtual host is down or broken. This can be very useful when you're running some test machines which can be broken in some way.
  • Have a visible and transparent system of resource distribution.

[su_quote] This can be appealed by BIG guys from BIG data centres, which can say that vCenter has a better distribution bla-bla... But I've seen what is going on in my system, those shitty pools can't keep crazy machines in their's stall!

When one of them, that tst crazy VM, starts to eat resources - nothing can stop it, and ESXi hangs, and this ESXI also can eat a lot of them (resources).[/su_quote] 

These are the first expressions from Hyper-V, they may differ from yours and I also can change them over time. but the first expression is still better than disappointment. Will see.