MOBILE AGENTS IN INSECURE ENVIRONMENTS
Franz Gruber
Software Competence Center Hagenberg
Austria
Abstract
The area of software agent technology is a heavily growing domain.
Already in the year 1992 describes agent based computing as the
next significant breakthrough in software development, and predicts
agent technology as "the new revolution in software".
Although these opinions might be exaggerated, currently, agents
are the focus of intense interest in many sub-fields of computer
science and artificial intelligence. Let us now take a brief look
at the most promising characteristics that make agent technology
so interesting for the research area of database technology.
An agent is mobile if it is able to migrate from one environment
to another. The migration of mobile agents might be strong or
weak respectively, depending on whether or not the agent is able
to store its internal execution state (i.e. instruction pointer
and stack), migrate, and continue its work at any internal point
of execution. An agent is able to exchange messages with the environment
and especially with other agents. Furthermore, collaboration requires,
that an agent is able to jointly work with other agents in addition
to simply communicating. An agent is called deliberative, if it
has an internal reasoning model empowering it to model its environment,
plan actions, and negotiate with other agents.
As agent platforms are expected to work in a highly insecure environment,
they have to implement a strong security system. DBagents can
adapt this security system and use it for authentication and authorization.
Biography
Franz Gruber (franz.gruber@scch.at) is project manager at the
Software Competence Center Hagenberg. Having graduated from the
Johannes Kepler University in Linz, he currently worked on several
research projects that investigate the applicability of agent
technology and other mobile distributed application architectures.
|
