An Immunologically-Inspired Adaptation Mechanism for Evolvable Network Applications

Provided by: University of Maryland University College
Topic: Mobility
Format: PDF
Large-scale network applications are expected to be more autonomous and adaptive to dynamic changes in the network to improve user experience, expand applications' operational longevity and reduce maintenance cost. Based on the observation that various biological systems have already met the requirements (i.e., autonomy and adaptability), this paper describes a biologically-inspired framework, called iNet, to design autonomous and adaptive network applications. iNet is designed after the mechanisms behind how the immune system works. iNet models a set of environment conditions (e.g., network traffic and resource availability) as an antigen and a behavior of network applications (e.g., migration and reproduction) as an antibody.

Find By Topic