Over the years, I have been involved with several global deployments of new technologies. The single common element I can identify among them is a stark disconnect between the field requirements perceived by corporate headquarters and how the territories themselves see their priorities. The topic of this blog is just one additional example of a dangerous pattern.
Corporate headquarters is usually a strong champion of datacenter consolidation. Not only does consolidation reduce overall costs, but it makes it easier to oversee and process data and services. The enthusiasm from the branch office is often more reserved. The branch manager has a completely different vantage point, which comes with a distinct set of concerns.
The idea of consolidation is great in principle, but what happens if my network goes down? Will my users still be able to access their applications within a reasonable response time? How will we make sure that all the data is backed up properly? Will I need more bandwidth and a more resilient Internet connection? How much will all this cost me?
These are all legitimate concerns. After all, centralization leads to loss of regional control. And a poor design will almost certainly have a negative impact on performance and availability. But all hope is not lost.
The fact that the data is stored centrally doesn’t preclude it also being replicated locally. The key is to effectively synchronize all information. A well-designed cache will not only match the user-perceived performance of a local implementation, but it will actually increase redundancy and therefore the overall availability of the service.
Let’s take content distribution as an example. If users need to share a fully loaded network uplink for every document they access in a central repository, they may find the experience frustrating. And their productivity would suffer if the connection is intermittent or prone to failure. But if all the content is replicated to the branch office — using a technology like BrancheCache — then users will have instant access to the local copies of their files. Any changes, whether they originate from the branch or the datacenter, will be replicated in the background to make sure the content remains synchronized.
Replication is also critical for business continuity. Whether it is a solution like BrancheCache or virtualization redundancy technology, such as Hyper-V Replica, the simple fact that data, applications, and state information are copied to distinct physical locations makes them less susceptible to outages and makes them an additional option for disaster recovery.
These are just some of the options for branch offices. Individually, they won’t necessarily apply to every branch or even every organization. But what all branches have in common is a separate set of requirements from headquarters. Usually, they will focus on the local implications of data availability and performance.
A well-designed architecture will embrace these needs with technologies that can address them. In fact, the transformed datacenter can actually improve the entire experience for the branch office. The key point to keep in mind here is that successfully transforming the datacenter means that the needs of all its users are met, wherever they are located.