VMware Cloud Community
rickardnobel
Champion
Champion

Reason for FT-enabled disk for FT VM?

If using Fault Tolerance the virtual disk for the VM needs to be FT enabled, i.e. eager zeroed. I am wondering what the reason for this is?

Would it be that the FT enabled VM needs that little bit higher disk IO or is it any other reason?

My VMware blog: www.rickardnobel.se
0 Kudos
7 Replies
vmroyale
Immortal
Immortal

Hello.

If using Fault Tolerance the virtual disk for the VM needs to be FT enabled, i.e. eager zeroed. I am wondering what the reason for this is?

Would it be that the FT enabled VM needs that little bit higher disk IO or is it any other reason?

Eager zeroed disks are simply the best performing disk type, as there is no overhead for writes.

Good Luck!

Brian Atkinson | vExpert | VMTN Moderator | Author of "VCP5-DCV VMware Certified Professional-Data Center Virtualization on vSphere 5.5 Study Guide: VCP-550" | @vmroyale | http://vmroyale.com
0 Kudos
rickardnobel
Champion
Champion

So there is just the extra performance of the eager-disk that is interesting, no other technical issue?

My VMware blog: www.rickardnobel.se
0 Kudos
vmroyale
Immortal
Immortal

As far as I know, the performance factor is it.

Brian Atkinson | vExpert | VMTN Moderator | Author of "VCP5-DCV VMware Certified Professional-Data Center Virtualization on vSphere 5.5 Study Guide: VCP-550" | @vmroyale | http://vmroyale.com
AntonVZhbankov
Immortal
Immortal

>Eager zeroed disks are simply the best performing disk type, as there is no overhead for writes.

To be more exact: there is no overhead for writes from the very beginning, but situation is absolutely the same for "thick" disks, without zeroing. So we can add here "security" reason.


---

MCITP: SA, MCTS Hyper-V, VCP 3/4, VMware vExpert

http://blog.vadmin.ru

EMCCAe, HPE ASE, MCITP: SA+VA, VCP 3/4/5, VMware vExpert XO (14 stars)
VMUG Russia Leader
http://t.me/beerpanda
rickardnobel
Champion
Champion

Thanks for your answers. I was just curious since it is a requierment for using FT.

My VMware blog: www.rickardnobel.se
0 Kudos
imomin
Enthusiast
Enthusiast

Eagered Zeroed thick disks are needed because if they are thin or thick, there will be a lot of logging data to be send from primary to secondary VM. So to avoid the logging link from getting saturared, we commit the entire provisioned space and zero it. This is for performance reasons.

0 Kudos
rickardnobel
Champion
Champion

Eagered Zeroed thick disks are needed because if they are thin or thick, there will be a lot of logging data to be send from primary to secondary VM. So to avoid the logging link from getting saturared, we commit the entire provisioned space and zero it.

Why would the zeroing be sent over the FT logging network? Is not the zeroing done by the vmkernel and the guest OS does not know it? Is that kind of low level "behind-the-scenes" commands sent over the FT network?

My VMware blog: www.rickardnobel.se
0 Kudos