Applying scale-out architectures to SQL Server

With an understanding of the source of big data, the uses for big data, and the technology for processing big data, there are changes and enhancements that are required for SQL Server to work in a scale-out environment. While SQL Server has long provided a parallel mechanism to scale processing within an instance, it requires new technologies to work with massive distributed datasets.

To scale out your computing environment, you need multiple computers, or nodes. But, of course, adding and removing physical computers from the ecostructure is expensive and time-consuming, not to mention complicated.

"Virtual" computers are a better solution. A technology called a hypervisor uses software to represent the physical computer's CPU, I/O, memory, and networking components to an operating system's installer, running an entire computer within a computer. The operating system installs onto a file that acts as a hard drive...