Systems and methods for fast detection of elephant flows in network traffic
Abstract
In a system for efficiently detecting large/elephant flows in a network, the rate at which the received packets are sampled is adjusted according to a top flow detection likelihood computed for a cache of flows identified in the arriving network traffic. After observing packets sampled from the network, Dirichlet-Categorical inference is employed to calculate a posterior distribution that captures uncertainty about the sizes of each flow, yielding a top flow detection likelihood. The posterior distribution is used to find the most likely subset of elephant flows. The technique rapidly converges to the optimal sampling rate at a speed O(1/n), where n is the number of packet samples received, and the only hyperparameter required is the targeted detection likelihood.
- Inventors:
- Issue Date:
- Research Org.:
- Reservoir Labs, Inc., New York, NY (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1805526
- Patent Number(s):
- 10924418
- Application Number:
- 16/270,089
- Assignee:
- Reservoir Labs, Inc. (New York, NY)
- DOE Contract Number:
- SC0011358
- Resource Type:
- Patent
- Resource Relation:
- Patent File Date: 02/07/2019
- Country of Publication:
- United States
- Language:
- English
Citation Formats
Gudibanda, Aditya, and Ros-Giralt, Jordi. Systems and methods for fast detection of elephant flows in network traffic. United States: N. p., 2021.
Web.
Gudibanda, Aditya, & Ros-Giralt, Jordi. Systems and methods for fast detection of elephant flows in network traffic. United States.
Gudibanda, Aditya, and Ros-Giralt, Jordi. Tue .
"Systems and methods for fast detection of elephant flows in network traffic". United States. https://www.osti.gov/servlets/purl/1805526.
@article{osti_1805526,
title = {Systems and methods for fast detection of elephant flows in network traffic},
author = {Gudibanda, Aditya and Ros-Giralt, Jordi},
abstractNote = {In a system for efficiently detecting large/elephant flows in a network, the rate at which the received packets are sampled is adjusted according to a top flow detection likelihood computed for a cache of flows identified in the arriving network traffic. After observing packets sampled from the network, Dirichlet-Categorical inference is employed to calculate a posterior distribution that captures uncertainty about the sizes of each flow, yielding a top flow detection likelihood. The posterior distribution is used to find the most likely subset of elephant flows. The technique rapidly converges to the optimal sampling rate at a speed O(1/n), where n is the number of packet samples received, and the only hyperparameter required is the targeted detection likelihood.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2021},
month = {2}
}