Milt Epstein wrote:
>
> On Wed, 22 Nov 2000, Myriam Abramson wrote:
>
> > Hi!
> >
> > Is there a difference in the performance of a servlet using http
> > requests vs. a server-side application using tcp/ip sockets?
>
> Not a real specific and/or complete answer (but then neither is your
> question :-), but I'd suspect that there would be, with the sockets
> winning.  It makes sense, since with sockets, you'd have all the
> control, and could fine-tune things for performance
>

 Just to play devil's advocate :-), you might want to
consider that while theoretically you might be able to
come up with a faster protocol, in reality you are
unlikely to be able to duplicate the years that http
server authors have spent fine-tuning their
implementations.

 Plus, in many cases involving servlets, the overhead
associated with the protocol is overshadowed by the
expense of whatever processing is required to generate
the response.

 So the protocol may not be a bottleneck, and if it's
not a bottleneck, you shouldn't waste time trying to
optimize it.

 But Milt is right, without more information there's
no way to give a proper answer...


-cks

___________________________________________________________________________
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff SERVLET-INTEREST".

Archives: http://archives.java.sun.com/archives/servlet-interest.html
Resources: http://java.sun.com/products/servlet/external-resources.html
LISTSERV Help: http://www.lsoft.com/manuals/user/user.html

Reply via email to