World Wide Distribution Point Replication Latency

[Lhsachs]

As I move to replicating my Master Distribution Point to the new world-wide distribution points on Windows Server VMs, I'm noting how long it is taking to replicate 20 GBs. Have any of you found ways to speed this process up?

I have Casper Admin Preferences for synching set to File Size.

Also, have any of you created a good means to test if there aren't Macs in some of the locations yet?
I've forced pushed packages back here from a distant server as a test...

Thanks in advance.

[jhbush]

It can definitely take a long time to sync 20GB of data depending on the pipe you have to work with and if the network that VM's are on is being throttled. I have DP's in lots of major cities around the world. They are hosted on mini's at the moment, but syncing 60GB to Sydney can take an entire day or more on the weekend. It will take longer when I move to the VM as our SCCM VM architecture is throttled and bandwidth monitored. In my experience DMG file sync much faster than say .pkg or .mpkg to SMB shares. Figured that out with the new CS6 programs.

[bentoms]

I worked with the Infrastructure teams to find a "sync window" when I could sync my servers during a time when a I would have the most available bandwidth across the MPLS.

Once we did that, I set my Mac Mini's to rsync Casper, NetBoot & ASUS data starting at the beginning of the window (they are actually staggered).

Sure it takes ages to sync from London to Hong Kong over the MPLS, but this way I know it's doing it at least.