W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1995

Effects of Persistent Connections

From: Mike Cowlishaw <mfc@vnet.ibm.com>
Date: Sun, 12 Nov 95 19:04:17 GMT
Message-Id: <199511121907.AA024173235@hplb.hpl.hp.com>
To: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Well, it looks as though enough clients are around for this to get
interesting.  For example, my server is now showing around 1% more
requests than transactions; all of the excess must be requests on
persistent connections.  I've also noticed a measureable increase in
throughput and an average response time drop.

It's interesting to note that the number of packets transmitted has
increased slightly (per request) from before; doubtless because
persistent connections require that TCPIP_NODELAY be set.  Has anyone
done any simulation of this side-effect?

(For live statistics on some of the above, try
http://www2.hursley.ibm.com/!statistics)

Mike Cowlishaw
IBM Fellow
Received on Sunday, 12 November 1995 11:12:01 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:35 EDT