add tool to asynchronously replicate across data centers

```
Current data replication is done synchronously.

There are requests to replicate data to backup weed file systems.

The tool should run continuously or on-demand. It should keep all the data in 
sync, including volume ids, file ids, and of course file data.

```

Original issue reported on code.google.com by `chris...@gmail.com` on 10 Apr 2014 at 8:44
