Debunking myths: proxies impact performance
In a recent blog entry Marc Logemann touches on the subject of proxy performance. In his entry he asks for a white paper by 'the Spring guys'. I don't want to spend (p)ages and (p)ages on discussing the differences up to the nanosecond between proxies and byte code weaving mechanisms, but I do think it's valuable to re-iterate once again what the differences are and whether or not this discussion matters at all.
What are proxies and why do we use them?
Let's first shortly revisit what proxies are used for (in general, and in Spring). According the Gang of Four (GoF) book on Design Patterns a proxy is a surrogate object or placeholder for another object to control access to it. Because the proxy sits in between the caller of an object and the real object itself, it can decide to prevent the real (or target) object from being invoked, or do something before the target object is invoked.In other words, proxies can be used as stand-ins for real objects to apply extra behavior to those objects--be it security-related behavior, caching or maybe performance measurements…