Detail publikace

A Fast FPGA-Based Classification of Application Protocols Optimized Using Cartesian GP

Originální název

A Fast FPGA-Based Classification of Application Protocols Optimized Using Cartesian GP

Anglický název

A Fast FPGA-Based Classification of Application Protocols Optimized Using Cartesian GP

Jazyk

en

Originální abstrakt

This paper deals with design of an application protocol classifier intended for high speed networks operating at 100 Gbps. Because a very low latency is the main design constraint, the classifier is constructed as a combinational circuit in a field programmable gate array. The classification is performed using the first packet carrying the application payload. In order to further reduce the latency, the circuit is optimized by Cartesian genetic programming. Using a real network data, we demonstrated viability of our approach in task of a very fast classification of three application protocols (HTTP, SMTP, SSH).

Anglický abstrakt

This paper deals with design of an application protocol classifier intended for high speed networks operating at 100 Gbps. Because a very low latency is the main design constraint, the classifier is constructed as a combinational circuit in a field programmable gate array. The classification is performed using the first packet carrying the application payload. In order to further reduce the latency, the circuit is optimized by Cartesian genetic programming. Using a real network data, we demonstrated viability of our approach in task of a very fast classification of three application protocols (HTTP, SMTP, SSH).

BibTex


@inproceedings{BUT119801,
  author="David {Grochol} and Lukáš {Sekanina} and Martin {Žádník} and Jan {Kořenek}",
  title="A Fast FPGA-Based Classification of Application Protocols Optimized Using Cartesian GP",
  annote="This paper deals with design of an application protocol classifier intended for
high speed networks operating at 100 Gbps. Because a very low latency is the main
design constraint, the classifier is constructed as a combinational circuit in
a field programmable gate array. The classification is performed using the first
packet carrying the application payload. In order to further reduce the latency,
the circuit is optimized by Cartesian genetic programming. Using a real network
data, we demonstrated viability of our approach in task of a very fast
classification of three application protocols (HTTP, SMTP, SSH).",
  address="Springer International Publishing",
  booktitle="Applications of Evolutionary Computation, 18th European Conference",
  chapter="119801",
  doi="10.1007/978-3-319-16549-3_6",
  edition="Lecture Notes in Computer Science",
  howpublished="print",
  institution="Springer International Publishing",
  year="2015",
  month="march",
  pages="67--78",
  publisher="Springer International Publishing",
  type="conference paper"
}