I’ve written several times about application nearness, and use the diagram below to highlight the fact that applications simply want to store data and get it back again.
In the 1980s, most applications were geographically “near” their data. Technologies such as Storage Area Networks (SANs) began to stretch this distance, and today’s mobile, scale-out architectures have driven the application even further away from its data.
Many data center architects have struggled to cope with this distance for a number of reasons (e.g. performance and security).
During a recent keynote at EMC Forum Australia, EMC CTO John Roese spoke about the “near to far” phenomena that has occurred over the last few decades, and issued a warning to the audience that applications are going “near to far and back again”.
There are a set of enterprise applications driving latency requirements that are pulling applications closer to their data. The diagram below highlights this reality.
What are the applications that are leading this charge? Here are a few:
How can a data center architect create an environment that supports this wide range of application nearness?
Roese hinted that the answer lies in the creation of a converged infrastructure (CI) strategy that is designed from the ground up to handle disparate workload distances. In upcoming posts I will begin to highlight several different design points that will move the industry towards a solution.
Before discussing CI, however, it’s worth pointing out that the diagram above does not represent the full spectrum of application distance.
In my next post I’ll cover a further subset of applications that move the bar even further to the right.