That seems pretty awesome, although unfortunately it's not in a stable kernel/iproute2 yet. Hopefully it will be very soon. I may still see if I can get it working in our environment, as that seems like it may be the best solution I've found as of yet. Alternatively, I'm looking at doing several HTB leaf queues (one for each general category of traffic), which would each contain an SFQ queue. On Sat, Jun 16, 2012 at 1:51 PM, Kenyon Ralph <kenyon@xxxxxxxxxxxxxxx> wrote: > On 2012-06-15T13:04:04-0700, Nick Bartos <masterchicken@xxxxxxxxxxxxxxx> wrote: >> That sounds like a nice idea, although it looks like only some of them >> support it (and I'm not sure if DCB requires support in the >> interconnecting switches/routers as well). I would really like to >> come up with a single solution I can use in all hardware cases. > > Have you been following the development of codel? Maybe this will work > for you? > > http://www.bufferbloat.net/projects/codel/wiki/Wiki/ > http://queue.acm.org/detail.cfm?id=2209336 > https://gettys.wordpress.com/2012/05/08/fundamental-progress-solving-bufferbloat/ > https://gettys.wordpress.com/2012/05/22/a-milestone-reached-codel-is-in-linux/ > > -- > Kenyon Ralph -- To unsubscribe from this list: send the line "unsubscribe lartc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
