as per the vsphere replication architecture whatever the changed block information that is sent by vsphere replication agent are captured by vsphere replication server in the form of redo logs and once the all the blocks are captured then redo logs gets collapsed but in MPIT they are maintaining it. Now my query is what is the size of redo logs in both the cases with MPIT and without MPIT.
It depends on the size of your VMs, their change rate, how many points in time you are keeping, how far apart, etc. There is no formula for this as there are too many variables.
If you want to get an idea, take an average VM for your environment (change rate, size, etc), set the MPIT where you want, run it for the duration (eg. if you are doing 6 points per day for 4 days, you'll need to wait 4 days) and see how big all of the snapshots are at the recovery site.
Does this answer your question?
When using MPIT the redo log size doesn't change compared with without because the MPIT state is created/maintained by use of VM snapshots at the recovery site.
Does this answer your question?
thanks for the reply. Generally what is size of this file
The size of the snapshot file? That will depend on the size of the VM, rate of change, number of points in time maintained and how frequently new points in time are created.
this is what i was assuming but when i saw this article at vmware blog i got confused as in the screenshot it is only shown as few MB. Can you please check this article and correct me if i am wrong. Site Recovery Manager with vSphere Replication Multi-Point-in-Time | VMware vSphere Blog - VMware Bl...
I'm sorry I'm not clear, what is your question? The VM in that screenshot has a low change rate so the snapshots aren't very large. The size of your snapshots will depend on what I outlined earlier.
ok. so in this do we need to plan for the double disk space in recovery site for an example?
It depends on the size of your VMs, their change rate, how many points in time you are keeping, how far apart, etc. There is no formula for this as there are too many variables.
If you want to get an idea, take an average VM for your environment (change rate, size, etc), set the MPIT where you want, run it for the duration (eg. if you are doing 6 points per day for 4 days, you'll need to wait 4 days) and see how big all of the snapshots are at the recovery site.
Does this answer your question?
Yes, Thanks again for your reply.