Lack of TCP self-clocking in practical networks

• Previous message: Kacheong Poon: "RE: Burstiness WIth increased max cwnd."
• In reply to: Kacheong Poon: "RE: Burstiness WIth increased max cwnd."
• Next in thread: Craig Partridge: "Re: Lack of TCP self-clocking in practical networks"
• Next in thread: Julien Godard: "RE: Burstiness WIth increased max cwnd."
• Reply: Craig Partridge: "Re: Lack of TCP self-clocking in practical networks"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hello All,

Kindly give me your opinions about lack of self-clocking in TCP.

I feel that TCP's self-clocking as illustrated in Van Jacobson's 1988
paper exists only if bottleneck link is a link of lower capacity (i.e.
bandwidth) than other links. However, if bottleneck link has higher
capacity than other links (i.e. the link is bottlenecked because a large
number of flows are using it simulateneously), there is no self-clocking
any more. I guess this is also a main reason why one needs some thing like
paced TCP. (The other reason being significantly less buffer
requirements with paced TCP.)

Thanks,
Mukul

• Next message: Craig Partridge: "Re: Lack of TCP self-clocking in practical networks"
• Previous message: Kacheong Poon: "RE: Burstiness WIth increased max cwnd."
• In reply to: Kacheong Poon: "RE: Burstiness WIth increased max cwnd."
• Next in thread: Craig Partridge: "Re: Lack of TCP self-clocking in practical networks"
• Next in thread: Julien Godard: "RE: Burstiness WIth increased max cwnd."
• Reply: Craig Partridge: "Re: Lack of TCP self-clocking in practical networks"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Sat Feb 26 2000 - 13:49:57 EST