The article considers the multicast-to-some generalization of thetraditional multicast paradigm. In this form of communication, amulticast group is still associated with each multicast message. It is,however, only necessary for any randomly determined subset of themulticast group to receive a multicast message. This subset is notdefined a priori nor is it necessarily the same from one multicast tothe next for the same group. It explores the use of this“probabilistic” multicast communication paradigm to addressscalability issues in the response collection application which arisesin some distributed computing applications. It also shows how this formof multicast can be used to reduce the discarding of responses to themulticast message due to buffer overflow at the multicast source
展开▼
