Article:		Nukes: the Open Source Java CMS
Subject:		Confused
Date:		2003-06-06 20:02:40
From:		anonymous2
I am confused by the author's point that the PHP implementation accessed the database for each page. He points out that this was a flaw in the design. Yet, doesn't the new Java implementation do the same thing? If not, how? And...why couldn't this be done in PHP? Having worked in both Java and PHP, I need more info in order to credit Java with scalability in this case. Sorry, but I smell a rat.

Showing messages 1 through 7 of 7.

• Confused
  2003-06-07 04:53:40  anonymous2 [View]
  
  
  I suppose the Java implementation just caches data instead of accessing the database all the time. This could be done in PHP as well, but the developers of PostNuke apparently have choosen not to.
  
  Hence you can credit the scalability improvement to the java developers instead, java developers are always smarter, ya'know ... ;-)
• Confused
  2003-06-07 20:30:58  anonymous2 [View]
  
  
  So how could one cache, for example, user login information in PHP so that you don't have to look it up in the db each time a page is accessed? For example, suppose you display the person's name and other such info on each page...if he's logged in.
  
  Thanks!
• Confused
  2003-06-07 21:17:32  anonymous2 [View]
  
  
  Store in a file-based session. This ties you to one machine, instead of being 'web-farm' ready, but it works. Dunno if it's how jboss/nukes is doing it or not though.
• Confused
  2003-06-07 21:45:08  anonymous2 [View]
  
  
  we store the user in memory, so it is lightning fast.
  
  accross a cluster we can use a replicated cache. clustered sessions in JBoss use it. that means that when the user is in memory and nukes updates it, the cache detects the field change (by bytecode manipulation) and replicates the change accross the cluster.
• Confused
  2003-06-09 05:31:30  anonymous2 [View]
  
  
  Is there a way to store user info, etc in memory using PHP?
• Confused
  2003-06-08 08:26:09  anonymous2 [View]
  
  
  So, let's say I've hit machine A in a cluster of 50 machines. What chance is there of me hitting machine J in the same cluster before the session data is replicated to that machine? .1 seconds? .01 seconds? I still have a hard time believing that *all* memory data (figure 500 users at 1k of session data each = 500k) can be replicated across a few dozen machines without there ever being the chance of data not being in sync. It seems like a lot of gyrations to do to get around a central database (which would also help survive power cuts, and which also gives an easier central place to do analysis of user session data from).
  
  BTW, the reason the *nuke clones are not scaleable is less to do with hitting the database on every request as it is that they don't know when to use the database and when not to.
• Confused
  2003-12-22 10:26:44  anonymous2 [View]
  
  
  Are you also saying that "clustering cannot not work". My be you should phone up Sun, BEA and IBM and tell them the bad news...