| Abstract |
The increasing popularity of Web-Services has exemplified the need
for scalable and robust discovery mechanisms. Although decentralized
solutions for discovering Web-Services promise to fulfill these
requirements, most solutions make quite limiting assumptions
concerning the number of nodes, the topology of the network and
a-priori information on the data(e.g. categorizations or popularity
distributions). In addition, most solutions are tested via
simulations using artificial datasets. In this paper we present a
lightweight, scalable and robust WSDL discovery mechanism based on
real-time calculation of term popularity. Results based on a
large-scale emulation on the DAS-3 distributed supercomputer, using
real data from Seekda.com show that we can achieve web-scale service
discovery based on simple keyword search. |