On Wednesday 17 August 2005 04:50 pm, Daniel Phillips wrote: > Hi, > > I've found that with the GigE interfaces I have here, MTU=9000 improves > throughput by about 31% versus MTU=1500. I run a fast network for my > cluster (crossover cable actually) bridged to the outside world over an old > card I had lying around that only supports MTU=1500. With the bridge up, > the local traffic falls back to the MTU=1500 speed. The reason is that the > bridge setup calculates the minimum MTU of all the bridged interfaces and > sets that as the bridge's MTU. > > One solution is to use a more modern card for the outside interface. But > is there any fundamental problem with fragmenting the bridge traffic when > necessary, and let the bridge MTU be the maximum of the interfaces instead > of the minimum? Is this a case of just didn't get around to it? > > Once again I plead lack of familiarity with the network code, but it seems > like the machinery needed to fragment packets in the forwarding path is > mostly already there.
1) Bridging occurs at the link layer (ethernet); fragmenting occurs at the network layer (IP). 2) A lot of protocols set the Don't Fragment bit, so you can't always fragment anyway. What you might want to do is set it up as an IP router rather than a bridge. That way, it will be able to fragment, or send ICMP responses when it can't. -John - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
