>> 6. Doing traffic shaping at the network edge is better than on the host node for > The host *is* the edge of the network. I'm sorry to have not mentioned that I consider the host nodes, or the end nodes, are not edges but instead something attaching on network edges. I consider the very last hub, or the access router which the end nodes connected to as the 'network edge'. And I'm sorry not to mention that practically it is probably better to do traffic shaping at the service provider network edge, for some reason of accouting, congestion control, or required by service-level-agreement (I don't think throwing bandwidth at the QoS issue is a good choice in the near foreseeable future, especially for the enterprise users.).
- Re: An alternative to TCP (part 1) stanislav shalunov
- Re: An alternative to TCP (part 1) Rahmat M. Samik-Ibrahim
- Re: An alternative to TCP (part 1) Randall R. Stewart
- Re: An alternative to TCP (part 1) Fred Baker
- RE: An alternative to TCP (part 1) Christian Huitema
- RE: An alternative to TCP (part 1) Larry Foore
- RE: An alternative to TCP (part 1) Tony Hain
- RE: An alternative to TCP (part 1) Bernard Aboba
- Re: An alternative to TCP (part 1) stanislav shalunov
- RE: An alternative to TCP (part 1) Jun'an Gao
- Re: An alternative to TCP (part 1) Jun'an Gao
- Re: An alternative to TCP (part 1) Jun'an Gao
- Re: An alternative to TCP (part 1) Jun'an Gao
- RE: An alternative to TCP (part 1) Jun'an Gao
- RE: An alternative to TCP (part 1) Larry Foore
- Re: An alternative to TCP (part 1) Keith Moore
- Re: An alternative to TCP (part 1) Jon Crowcroft
- Re: An alternative to TCP (part 1) Colin Perkins
- Re: An alternative to TCP (part 1) Jun'an Gao
- Re: An alternative to TCP (part 1) Harald Alvestrand
- Re: An alternative to TCP (part 1) Jun'an Gao