Machine learning is becoming a buzzword—arguably an overused one—among companies that deal with networking. Recent announcements have touted machine learning capabilities at Google, Hewlett Packard Enterprise (HPE), and Nokia, for instance.
But machine learning isn’t being applied to networking itself. Why is that?
The intersection of machine learning and networking is where David Meyer, chief scientist at Brocade, has been working. After serving a term as the first chairman of the OpenDaylight Project’s Technical Steering Committee (TSC), Meyer shifted his work into the realm of artificial intelligence.
Even though networking has “just massively more compute and massively more data” available, it’s not yet clear how machine learning can be applied there, Meyer says.
What’s missing, he believes, is a theory of networking.
A rich body of academic work backs the networks we use today, certainly, but there is no unifying theory defining how a network, in an abstract sense, ought to behave, or how it ought to be structured. The networks that form the Internet certainly share some core principles, but they weren’t built from a central theory. They emerged through trial-and-error, “some good ideas and people telling each other how to do it,” Meyer says.
Machine learning, on the other hand, “is just math,” he says, and math requires models.