CHAPTER 4 Data Networks—The Nervous System of the Cloud

This chapter picks up just at the point where Chapter 3 left off. There we had already mentioned data networking more than once, always with a forward reference. No postponement anymore!

Data networking refers to a set of technologies that enable computer-to-computer communications. The ultimate result is that two processes located on different computers can talk to one another. This, in turn, supports distributed processing. As you may remember, this is an operating system task to enable interprocess communications on the same machine. Transferring this capability across machines—over a network—naturally involves operating systems, too.

In fact, the disciplines of data communications and operating systems have been evolving side by side since the 1960s. In earlier systems, the data communication capabilities were added by means of both (a) new device drivers for physical network access and (b) specialized libraries for interprocess communications across the network (which differed from those provided by the kernel for interprocess communications within the machine). It is interesting that, in these early days, achieving over-the-network interprocess communications was not an end in itself1 but rather the means of accessing remote resources—typically, files. File transfer was a major data networking application. Another one was transaction systems (such as those employed in banking or airline reservations), in which a ...

Get Cloud Computing now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.