Navigation
User login
Quote
John von Neumann
It would appear that we have reached the limits of what it is possible to achieve with computer technology, although one should be careful with such statements, as they tend to sound pretty silly in 5 years.
(Said in 1949)
(Said in 1949)
The Flyweight Design Pattern
Tue, 06/15/2004 - 13:13 — kebernet
Good article on caching for J2EE apps at The Server Side
This work is licensed under a Creative Commons License.
1998-2008 temple of the screaming penguin
TotSP Search
Community
Chatter
- You don't need anything
1 sec ago - Hey nice article however, i
2 days 13 hours ago - Appreciation
4 days 9 hours ago - thanks for information...
6 days 6 hours ago - In total agreement on this one
2 weeks 17 hours ago - Suggested change to wrapper
2 weeks 4 days ago - Hmmmm...i jus called it and
2 weeks 4 days ago - I just quit!
2 weeks 4 days ago - Touche!
Pun intended.
2 weeks 6 days ago - Re-pricing the Touch
2 weeks 6 days ago
Comments
RE: The Flyweight Design Pattern
cool article, being such a novice insofar as formal education relating to programming goes i had never heard of Flyweight, very simple and obviously very effective.
But I would not really call it a "cache". To me, maybe I am incorrect, cache implies temporary and subject to changes that must be updated? This I guess could be called a "cache" but it seems that the canonical cache objects will just always be there, always the same, none of the complexities of update and such normally associated with caching.
RE: The Flyweight Design Pattern
Yeah. *REALLY* THe Flyweight is what persistence systems like JDO and Castor are centered around, as an "obvious practice" system. Its really more about maintaining your model in ram and hitting back to the datastore when you need to refresh it. I'm kinda usinging a halfassed version of this for the jTOTSP model layer, but its a little tricky since I don't want to use static datasets because I want someone to have the option to deploy multiple jTOTSPs in the same JVM.