Re: Effects of Persistent Connections

• From: Jeffrey Mogul <mogul@pa.dec.com>
• Date: Mon, 13 Nov 95 11:17:36 PST
• To: Mike Cowlishaw <mfc@vnet.ibm.com>, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
• Message-Id: <9511131917.AA27060@acetes.pa.dec.com>

Simon beat me to it:
   I don't really understand why TCPIP_NODELAY would need to be set for
   P-HTTP; although I haven't implemented the P-HTTP specifications, I
   have taken measurements for HTTP-NG operating in the closest
   available mode, using the HTTP-TOS relay facility.

While Simon then goes on to discuss why he doesn't think TCP_NODELAY
is required for HTTP-NG, he does leave open the more immediate issue
of whether it's required for persistent connections in HTTP V1.1.

I would not be entirely surprised if TCP_NODELAY were useful in
the context of persistent connections, but it's Monday morning
and I can't figure it out for myself.  So if someone could explain
in some detail why it is required, that would be nice.

Thanks
-Jeff

Received on Monday, 13 November 1995 11:51:23 UTC