Organizations who perform enterprise scale large deployments of SharePoint are very sensitive to any downtime of the system so that there is minimal affect to their employees and clients.
With the release of SharePoint 2013 RTM, Microsoft had changed their policy from SharePoint 2010 to not officially supported server farms that spanned across multiple datacenters.
Previous text taken from from the TechNet Hardware and software requirements for SharePoint 2013 article (Now outdated and should be updated soon – http://technet.microsoft.com/en-us/library/cc262485(v=office.15).aspx#hwLocServers):
Hardware requirements—location of physical servers
All servers that belong to a server farm, including database servers, must physically reside in the same datacenter. Redundancy and failover between closely located data centers that are configured as a single farm (“stretched farm”) is not supported in SharePoint 2013.
Stefan Gobner from the SharePoint team has recently published an article with an update on stretch farm support in SharePoint 2013.
I guess with some performance and capacity testing, Microsoft has gone back and changed it to officially announce limited supportability for stretched farms based on a prerequisite of minimal (< 1ms) latency between components of the farm.
Other things to consider for stretched farms from Stefan’s post:
Q1 My customer wants to distribute their topology across one or more distinct geographic boundaries (i.e. between cities, states, provinces), is this supported?
Q2 My customer maintains a logical datacenter comprised of one or more physical buildings on a single site. Is this supported?
A2 Yes, providing there is a highly consistent intra-farm latency of <1ms, 99.9% of the time over a period of ten minutes. (Intra-farm latency is commonly defined as the latency between the web front-end and database servers)
Q3 My customer’s latency exceeds 1ms – what can I do to get them to a supported configuration?
A3 See http://technet.microsoft.com/en-us/library/cc263031.aspx for documentation detailing high availability and disaster recovery topologies possible with SharePoint Server 2013.