Genomics in the AWS Cloud

CHAPTER 5AWS Services for Genome Analysis

Cloud computing is not a new concept. Ever since data networks have existed, there has been the idea of people here using computing resources there. Way back at the dawn of electronic computing history (the early 1970s), the standard way to do anything with a computer was to sit down at a dumb terminal (i.e., a device that had no computing power of its own but rather transmitted instructions to and received results from a remote computer) and interact with a machine from a distance—usually from somewhere else in the same building.

Later, say from the early 1980s until when the Internet really took off near the turn of the 21st century, we had what was called client-server computing. Under this arrangement, workplaces had machines (servers) that specialized in particular tasks, such as storing files that many people had to access, running a printer, or hosting a database. People sat at other computers, usually less powerful than the servers, called clients. The software on the clients interacted with the software on the servers, and while computational work was done on the client side as well as on the server side, the ultimate goal was usually to update data that resided on a server.

In the case of an office or other workplace, the servers usually resided in a special room in the same building or in another building nearby. Particularly critical servers—say, those that controlled the telephones and contact centers of large companies—were ...