Systems and methods of data flow classification

Abstract

Systems and methods of classifying data flows being communicated on a network by one or more network elements. One method includes creating a table including information of packet timestamps and pre-defined packet header fields, grouping packets into data flows based on information in the table, assigning flow identifiers to each data flow, determining a plurality of feature / characteristic sets having one or more features and / or one or more characteristics of the data flows, determining one or more classifiers to predict flow labels using the plurality of feature / characteristic sets, and generating a classification policy that includes the one or more classifiers to classify data flows on the network. The method can also include storing the classification policy in at least one non-transitory computer medium that is accessible by a network element that is classifying data flows on the network, and using the classification policy to classify data flows.

Description

INCORPORATION BY REFERENCE TO ANY PRIORITY APPLICATIONS[0001]This application is a continuation of U.S. application Ser. No. 16 / 694,117 filed Nov. 25, 2019, now U.S. Pat. No. 10,855,604, which claims the benefit of U.S. Provisional Application No. 62 / 771,635 filed Nov. 27, 2018, titled “DATA CENTER DATA FLOW CLASSIFICATION VIA MACHINE LEARNING” which is incorporated by reference herein in its entirety.[0002]Any and all applications for which a foreign or domestic priority claim is identified in the Application Data Sheet as filed with the present application are hereby incorporated by reference under 37 CFR 1.57.BACKGROUND OF THE INVENTIONField of the Invention[0003]This invention relates to transmission of digital information over a communications network. More particularly, this invention relates to characterization of data traffic flows in a network in real-time or near real-time.Description of the Related Art[0004]Over seventy percent of all Internet information traffic takes pl...

AI Technical Summary

Benefits of technology

[0007]The disclosed embodiments can provide a network element or network elements in a data center, or in other network, a predetermined classification policy that can include systems, processes and software to classify data flows as they are being transmitted through a network. For example, classifying data flows in real-time or near real-time as elephant flows and mice flows inside data centers or other networks, so that those elephant flows or mice flows can be subjected to special handling to improve the flow of all data in the data center.

Problems solved by technology

Currently, most data traffic management methods are either too high-level and rigid, or too granular and complicated.

Too granular and complicated methods include such data center protocols as equal-cost multi-path (ECMP) protocols, where routing decisions are made on each packet by mapping multiple data links with the same number of hops from source to destination and routing the packet along those multiple “equal cost” paths, or managing the buffers on network devices so that the “right mix” of traffic resides in the buffers and other traffic is excluded or discarded from the buffers.

Images