Specifications
Best Practices for Virtualizing and Managing Exchange 2013
28
28
To eliminate the inefficient and unnecessary steps required by traditional host-based file transfers, ODX
uses a token-based mechanism for reading and writing data within or between intelligent virtual storage
database volumes (Figure 16). Instead of routing the data through the host, a small token is copied
between the source and destination. The token serves as a point-in-time representation of the data. For
example, when you copy a file or migrate a virtual machine between storage locations, Windows Server
2012 copies the token representing the virtual machine file. This removes the need to copy the underlying
data between servers.
Figure 16: Offloaded Data Transfer in Windows Server 2012
The ESG Lab tested the efficiency and functionality of Offloaded Data Transfer. Two servers were
connected to an ODX-compliant Dell EqualLogic storage array.
24
The storage array consisted of 12 SAS
drives (600 GB each). A single RAID5 pool was created with two volumes: One contained a 75 GB virtual
machine, and the other was empty. Using an intuitive wizard, the ESG Lab configured a virtual machine
live migration from one server to another within a SAN. The lab specified the type of move, the server
receiving the data, move options, and destination virtual machine options. It then transferred a virtual
machine using the traditional non-ODX method and the new ODX method. The lab monitored network
utilization and elapsed time for the transfer to complete in both test cases.
The results in Figure 17 show noticeable improvements using ODX. The ODX transfer took approximately
6.5 minutes for the virtual machine to completely migrate to the other server, and the average network
bandwidth consumption was around 64 Kb/sec. Conversely, with non-ODX method, moving the 75 GB
virtual machine over the network took approximately 52 minutes and consumed 4 Mb/sec of network
bandwidth. The ODX method completed eight times faster than the non-ODX method, while consuming
virtually no server CPU or network resources.