[Live-devel] Multicast Address Allocation API

Ross Finlayson finlayson at live.com
Mon Apr 26 23:55:06 PDT 2004


>By the way, what is the status of RTT negative value issue?

Good news: I have been able to reproduce this problem myself, and am 
currently looking into the cause (and, I hope, a solution).  So stay tuned...

>Just out of the blue. I came across RFC 2771 (multicast address allocation).
>Do you think Live.com's RTP library will later implement multicast address
>client that allow dynamic multicast address instead of the prespecified one?

Probably not anytime soon.  The basic problem is that there are few dynamic 
multicast address allocation *servers* (i.e., MADCAP servers) deployed 
anywhere.  So there's no point in implementing the client side of the 
protocol, when there are few, if any, servers for it to talk to.

Instead, if - as is usually the case - your multicast application involves 
streaming from a single source only, then instead you can use a variant of 
IP multicast called "source-specific multicast" SSM.  If you're using SSM, 
then you can use the LIVE.COM routine "chooseRandomIP4SSMAddress()" to 
choose a multicast address.  (There are a few examples of this in the 
"testProgs" directory.)

If you want to learn more about Source-Specific Multicast, you can find 
several RFCs and Internet-Drafts using Google.  (Note: A detailed 
discussion of SSM would be largely off-topic for this mailing list.)


	Ross Finlayson
	LIVE.COM
	<http://www.live.com/>



More information about the live-devel mailing list