Why Data Gravity Will Grow Stronger

The term “data gravity” refers to the desire to have applications and data attract more applications and data on a network. The idea is based on Newtonian gravity: The larger a mass, the more the attraction. The term was coined by Dave McCrory in 2010.

One of the first things you discover with large data sets is that they are hard to move. The need for low latency and high throughput drives data gravity. AWS famously rolled out its “snowmobile” service to help customers with moving up to 100 petabytes of data per truck. It is literally a storage data center in a box delivered by a semitrailer truck. If you had a full 10 Gbps connection straight to the cloud that you could maximize throughput on, it would take nearly three years to transfer that much data. That’s the throughput problem in a nutshell.

Additionally, applications that want access to data want it to be fast. If you have an application that runs in a data center in Chicago and it needs to access data in a data center in Ashburn, Virginia, you have to deal with waiting for data to flow back and forth. If your application and data are both in Ashburn, you can access the data in 5% of the time. That can mean your application can work over 20 times faster for some cases. That’s the latency component of data gravity.

Big data is a rapidly growing market. The enormous market for big data solutions is driven by the value they can bring to enterprises. For example, McKinsey estimated a potential $100 billion in annual value for the health care system in the U.S. However, if you look at big data success stories, you will notice that a lot of big data projects seem to only require data from an organization. But there are efforts to unearth unstructured data that is not electronically readable, and there are a wide variety of efforts to safely share data between organizations.

Finish reading this article at Forbes