Support for offline replication data transfer from on premise to Azure
A feature that enables to faster, effective & secure export of large number of virtual machine replication storage using an external storage device and ship it to Azure for import.
Support will be delivered for migration scenarios. In DR scenarios customer bandwidth requirement for delta replication and initial replication are similar. Hence not planned for DR scenarios
I have been using Azure site recovery for almost 2 years now as a DR setup for 1 data centre. Due to increased disk usage, 1 of the VMs now has a disk of over 4095 GB. As a result I am not able to replicate this critical server and currently it is not possible to reduce the disk size in VMware. Is there a possibility for the disk size to be increased?
Mark Stevenson commented
We are currently in the process of trying to perform the initial replica of terabytes of data to Azure ASR over a range of 10mbps and 20mbps links which we have to throttle back to 50% during working hours. This is very slow and painful.
By breaking the replica to our (soon to be decommissioned) co-lo Hyper-V server so we can use Azure ASR, we are exposing the business to significant risk during the prolonged time it is taking to perform the initial replica. Providing a solution to reduce the initial seed time has to be high on the priority list for customers wanting to adopt Azure ASR for Hyper-V sites.
Nash, Michael commented
Bandwidth requirements for delta replication and initial replication are not similar at all. A large file server (3+ TB) with small delta changes would require a particularly large bandwidth for seeding in a reasonable time-frame where the deltas wouldn't require the same.
Allan Ruiz commented
Azure Data Box... Azure Data Box Disk... Azure Data Box Heavy... and no ASR support... You're missing an opportunity. Since 2014...
Brad Busch commented
This is available for Azure Backup already, let's get the teams together and make this available for ASR. This was being planned in 2015, what happened? There is obviously a use case for this.
This functionality is also needed when replicated VM's are moved between on premise datacenters. It would be nice to be able to keep the data already replicated and use it to preseed the VM replication after it has been moved to the other datacenter.
Brian “B” Laws commented
This feature is critically needed. My customer has one SQL Server VM with 17+ TB of disk and another with 8.5 TB with only a 1 Gbps uplink. We tried ASR protection but found that the storage array would run out of space, crashing the server. We would like to use ASR as the single solution for their entire DR, but large VMs like this make that impossible. Being able to seed huge VMs like this using offline replication could enable these scenarios.
Carlos Daniel commented
I'd like to add my voice to this request.
Bryan Rias commented
Looking at a customer with 125 Servers to be protected. The # of TB's that by default come with that # of VM's, would limit any efficient manner to replicate systems in waves. The waves get drawn out too long without seeding.
The Offline Replication Seeding with Data Box or other Snowball like service, knowing you want to go after larger customers, should have been road-mapped closer to MVP.
Please consider this capability, as you are causing creative variances from a viable ASR solution, that typically causes providers to chose other DR solutions that handle pre-seeding options out of the box. That way we can manage overall DR projects more efficiently.
When I look at customers that have Hyper-V and or Physical servers, it would be nice to have ASR as an unhindered lever or decision to make. Instead, we look at other options which is a shame.
We need this in SA for low bandwidth customers. Please!
We have a customer who would like to replicate a series of VMs with a total data capacity of over a Petabyte which makes using ExpressRoute challenging even at its highest capacity.
Benjamin Cramphorn commented
Initially requested in 2014 and only now is it being reviewed?
Sam Wozniak commented
Yep, initial replication of large VMs is the biggest PITA with ASR IMHO. I see the Import/Export service can do something, but I didn't see any ASR specific support in any doc I read recently.
Brian Supple commented
I'd like to add my voice to this request. It's a road block for ASR with many of our customers. The initial upload is taking too long and is hammering their internet connection.
Carlos Sanchez commented
There are some news on this? At least some hint..
This feature is needed, even by customers who have high bandwidth but do not want to wait weeks for the initial replication. This service would see considerably higher usage with this improvement. Any progress on it as of March 2017?
Ted Lo commented
This feature would help out so much!!
Jacky Zhou commented
ASR support Thanks, CU has more than 10 TB need to sync...