作者: Pavel Tichý
DOI:
关键词: Bottleneck 、 Scalability 、 Single point of failure 、 Computer science 、 Enhanced Data Rates for GSM Evolution 、 Multi-agent system 、 Fault tolerance 、 Distributed computing 、 Service provider 、 Knowledge engineering
摘要: Agents in a multi-agent system usually use middle-agents to locate service providers. Since one central middle-agent represents single point of failure and communication bottleneck the system, therefore structure is used overcome these issues. We designed implemented called dynamic hierarchical teams that has user-defined level fault-tolerance moreover fixed scalable. prove reruns size λ vertex edge connectivity equal λ, i.e. stays connected despite - 1 failures or channels. focus on social knowledge management describing several methods can be for propagation search this structure. also test practical experiments.