On Mon, 20 Oct 2003, Zhao, Forrest wrote: > Recently I found an IETF reliable multicast working group > site(http://www.ietf.org/html.charters/rmt-charter.html). > Since the kernel has the support for multicast now, do you > think it's necessary to support reliable multicast in kernel? > Are there many applications that need this rmt feature until now? Multicast stream or bulk data transfer is much less relevant today that it was a few years ago. Most of the appeal of multicast was that the underlying hardware implemented a shared broadcast media. Thus local multicast was almost free. But very few installations still use shared media or repeaters. Today's networks are implemented using switches, where multicast traffic has to be handled as an exception in the switch firmware. Thus using full-bandwidth multicast is very likely to result in packet loss. While it's possible to still get good performance with a specific configuration by selecting an algorithm that works well with the loss characteristics of the switches and the transmit patterns, this is not a general purpose, deployable solution. I'm not saying that multicast is in any way evil. It's wonderful for local service discovery and low-rate status reporting that would otherwise be done with broadcast packets. And, at a different level, multicast allow wide-area internet "broadcasts" with carefully managed networks of "repeater" hosts. But none of these uses indicates that we need reliable multicast implemented in the kernel. -- Donald Becker becker@scyld.com Scyld Computing Corporation http://www.scyld.com 914 Bay Ridge Road, Suite 220 Scyld Beowulf cluster system Annapolis MD 21403 410-990-9993 - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
