Originally Posted by
GregBrannon
Crossposted here.
What do you mean by "reliably." How will you measure reliability? Software can do what it has been programmed to do with 100% reliability, but there are many external factors that can impact the reliability of the communication that the sending software cannot control. The communication should be tested in the most stressful environment anticipated to determine overall reliability. Further, a multicast message usually indicates that there is no verification of receipt, so the sending software will not know whether all intended clients received the message. Without acknowledgment, the sending program won't know to resend.
If you provide more details about what you're trying to do, what concerns you have with the communication, and how you'll be measuring reliability, you may get more specific help or guidance.
Hi
By reliability ,I mean an ordered delivery of all packets to all the recipients of the group.I need to do this for my school project. I downloaded a normal multicasting program from internet in order to test it as to how the packets go missing. I executed the program by opening multiple clients(command prompts) on the same host. I see that all the packets arrive to all the clients in an orderly manner. Fistly I would like to know why is this happening, because I studied in my course that packets may or may not get delivered to all the clients in multicasting . Is it because I have multiple clients over the same host? Should I test it by putting clients in different hosts?
The main objective is to make ensure delivery of the packets across all the clients in multicasting(reliability), which I would be able to do only if I test in what way and in what circumstances packets go missing in multicasting.
(PS I have used Ex 3 program from website
Java Multicasting to test multicasting)
thanks