Two days ago, on a lazy evening, I started work on my first solid Jaxer project — building a memcached client. As a testimony to the elegance of JavaScript, it took me less than 2 hours to get a basic memcached client up and running.
What is memcached, you ask? From the memcached website:
memcached is a high-performance, distributed memory object caching system, generic in nature, but intended for use in speeding up dynamic web applications by alleviating database load.
What it does is, lets you use excess RAM on your server machines as temporary stores of data, so that you can avoid going to your DB when you want frequently accessed data. Needless to say, accessing data from RAM is much faster. Since the creation of memcached for LiveJournal.com, memcached has become the cheapest and fastest way to scale a website.
Now, my implementation is not feature complete, nor at peak performance, but I aim to take it there. Even so, I decided to get my code out and available for everyone to see, so that people can tell me what they think of it.
You can go to the Google code page for the JsMemcached-client project, or download the code right away. Though limited in features and with a possible bug in an edge case in the get call, you can start using this already.
This is my first open-source project, so please don't be too harsh in your feedback.
Help needed: I need to figure out how to performance-optimize this code. Either I need to get rid of that loop in the get call, or I need to write a wrapper around libmemcached. I would prefer the latter, since I have to add more features for feature-completeness. Any idea on how I can get Jaxer to talk with libmemcached (or any other such out-of-process thing)?
