Re: The future of ICP_OP_HIT_OBJ

rousskov@plains.NoDak.edu writes:

>On Tue, 18 Nov 1997, Duane Wessels wrote:
>
>> I'd like to get an idea of to what extent people think the ICP HIT_OBJ
>> feature is a good idea to continue supporting...
>
>Has anybody measured the actual improvement from using the ICP HIT_OBJ
>feature? That is, how often it happens, and how much response time is
>saved?

I just did a quick-n-dirty measurement which indicates that
ICP_OP_HIT_OBJ is only slightly faster thant HTTP/1.1 persistent
connections. The difference seems to be about one RTT, due to
the ICP query/reply before the HTTP request.

See http://ircache.nlanr.net/Cache/hit-obj-vs-http-1.1/ for a graph.

Given this, and the sentiments of others, I see very little reason to
keep the HIT_OBJ functionality in ICP.

Duane W.
Received on Wed Nov 19 1997 - 14:56:52 MST