Using multiple WAN accelerators for a Copy Job

[kjparker]

I have a copy job that mirrors my backup repository across a high bandwidth but latency link for offsite storage. When I use the WAN accelerator, the total data transferred across the wan is obviously reduced, however the entire data transfer takes considerably longer due to availability of wan accelerators.

Can I create a number of local wan accelerators, and do a "Many to one" wan accelerated copy job? It would appear that I am only able to use a single accelerator on each end. Is there something I have missed?

[HannesK]

Hello,
no, you did not miss anything. It's a known side-effect we are working on.

If you create multiple backup copy jobs, then you can use multiple WAN accelerators for "many-to-one". There is no general rule of how many "many-to-one" make sense with high bandwidth, so you need to play around a little bit on your own.

Best regards,
Hannes

[wishr]

Hi Keith,

Sure, here is the description of this setup. Global cache might help to reduce the amounts of data being transferred over WAN, however the network bandwidth on the target WAN might still be the bottleneck.

You can configure your copy jobs to be split between several source WAN accelerators while targeting a single target WAN. For example, Job1 is will be using WAN1 as a source and targeting WAN3, while Job2 will be using WAN2 targeting WAN3, etc. and give it a try.

Regards,
Fedor