Return-Path: Mailing-List: contact tomcat-dev-help@jakarta.apache.org; run by ezmlm Delivered-To: mailing list tomcat-dev@jakarta.apache.org Received: (qmail 10216 invoked from network); 16 Jun 2000 17:08:21 -0000 Received: from mail.singingfish.com (@216.254.3.225) by locus.apache.org with SMTP; 16 Jun 2000 17:08:21 -0000 Received: (qmail 29589 invoked from network); 16 Jun 2000 17:05:28 -0000 Received: from unknown (HELO c363190b) (10.0.0.62) by mail.singingfish.com with SMTP; 16 Jun 2000 17:05:28 -0000 Message-ID: <004a01bfd7b5$31934250$3e00000a@sttln1.wa.home.com> From: "Jonathan Reichhold" To: References: <001901bfd773$291adb20$6500a8c0@internal.digitalkiwi.com> Subject: Re: Global caching hack? Date: Fri, 16 Jun 2000 10:06:20 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 X-Spam-Rating: locus.apache.org 1.6.2 0/1000/N What about an RMI object that can be accessed from each servlet/JSP? This way you have one cache which is available from multiple JVM's and contexts and you can keep the pool to a single instance? Not sure of the symantics of the JSP page (may need to use a bean), but why not try it? Does anyone have any experience that would show that this would be a bad thing to do? It seems to me that this is a really powerful technique which isn't used nearly enough. Thoughts? Jonathan ----- Original Message ----- From: "Robin Giese" To: Sent: Friday, June 16, 2000 2:13 AM Subject: Global caching hack? > Hi, > > I was wondering how I'd be able to hack my own global cache (or rather, > bunch o' variables) into Tomcat that would be available to all servlets and > JSP pages. I'm trying to share an Oracle connection pool (and a couple > similar, global, very large, non-session dependent objects) to all my > servlets and JSPs, and I don't want to build a different cache for each > servlet and JSP page, because it would be an incredible waste of memory. > Where would be the right place for this kind of hack? Or is there an > existing facility I could hack up to make it not a complete hack? (However, > the objects aren't serializable.) > > Thanks, > > --robin > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org > For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org > >