The PHP::Impact blog has a recommendation for those using the memcache functions in their PHP applications - be sure your hashing strategy matches what your script does.

If you are using the Memcache functions through a PECL extension, you can set global runtime configuration options by specifying the values within your php.ini file. One of them is memcache.hash_strategy. This option sets the hashing mechanism used to select and specifies which hash strategy to use: Standard (default) or Consistent.

The recommendation is to set it to consistent to allow for the most flexibility on adding and removing servers from the caching server pool without the need for outside intervention.

On his blog today Gopal Vijayaraghavan has posted about the release of the latest version of the APC sofware (Alternative PHP Cache) - version 3.1.2.

Finally, after nearly a year of work, it's into a release. Some new stuff has sneaked into it undocumented, that people might find interesting - apc.preload_path would be one of them. The backend memory allocation has been re-done - the api part by me and the internals by shire. There's a hell of a lot of new code in there, both rewritten and added. Tons of php4 cruft removed, php5 stuff optimized, made more stable, then less stable, made faster, then applied brakes. Made leak-proof, quake-proof and in general, idiot-proof. So, on & so forth.

To show the difference, he includes a diff of the current version against the previous - 68 files changed, 3255 insertions and 5545 deletions.

In this new post to his blog, Brandon Savage looks at a few methods you can use to help reduce the overhead connecting to the database can cause.

Below are a list of my top five quick-and-dirty strategies for improving database performance in web applications. These suggestions are culled from recent experience and mixed with some ideas that I've implemented in my own code. They're not high level, but they are something we need consistent reminders about.

Here's the short version of the list:

• Try caching
• Reduce the number of queries that run
• Use indexes
• Optimize data usage
• Avoid functions in WHERE statements

You'll have to check out his post for the full versions, though.

On the Tillate Blog today there's a new post from Leo looking at how you can give clients back the control on when your cache refreshes.

We worked hard on the right caching strategy over months now. But there are still caches that do not get invalidated right or are cached too long and we still get complaints from our users. So we decided to integrate a feature that allows certain users to invalidate the cache on the server side. The first idea was to add a link to every page that will append a GET-parameter to the site and then avoid the cache.

He used the apache_request_header function to get the current request's headers and noticed that they included a cache-control header. By changing up this on the client-side, it allows the site visitor to "force" an update to the information if your application pays attention to it.

Chris Jones has written up another post about a feature in some of the latest OCI8 libraries for PHP - result caching.

Oracle Database 11g introduced "server-side"and "client-side" result caches. The database cache is enabled with the RESULT_CACHE_MODE database parameter, which has several modes. With the default mode, queries for which you want results to be cached need a hint added - No PHP changes are required.

This caching is perfect for results from smaller lookup tables and can drop CPU usage by quite a bit. Oracle automatically takes care of validating the cache entries when something changes the stored results so your query is always fetching the latest (without having to worry about a timestamp on the cache or anything like that). He gives a few examples of how it would work, comparing the queries both with and without the caching.

Brett send in a link to a new post on his "Mr PHP" blog about a method he's come up with to cache images using the phpThumb library.

Generate thumbs by visiting a URL such as your.com/thumbs/images/image.50x50.jpg. This will create a 50x50px thumbnail of your.com/images/image.jpg. The thumb will be stored on your server at your.com/thumbs/images/image.50x50.jpg so the next request for the same image will be loaded without loading php for ultra fast image cache.

The phpThumb tool lets you dynamically resize an image and, if one does not exist for it, make a thumbnail that's output and cached. From then on the script just pulls from that local copy. He includes his code to get it working and the mod_rewrite rule that maps an image request back to it (so it's still "/images/myimage.jpg" instead of "/app/phpThumb.php?src=myimage.jpg&w=100&h=100" in the img src).

In a recent post to his blog Dhiraj Patra looks at the caching functionality that PEAR has to offer via the PEAR Cache package.

Caching is currently a hot topic in the PHP world. Because PHP produces dynamic web pages, scripts must be run and results must be calculated each time a web page is requested, regardless if the results are the same each time. In addition, PHP compiles the script every time it is requested. [...] PEAR's Cache package offers a framework for the caching of dynamic content, database queries, and PHP function calls.

He talks a bit about what kind of methods are included with the package and shows examples of how it works for function call caching, caching the output from the script execution and how to implement your own custom caching extension of the main code to make it even more flexible.

Doug Brown has posted a new entry to his blog about how the Zend_Cache component of the Zend Framework has saved him some money.

He and his site were taken offline by his own hosting company because of one little detail - the limit on usage for the shared server was 3% and his site was using 30%. Going through his logs, he found the culprit - a "too many connections" message from MySQL due to the number of requests.

I'll admit, I was in a huge hurry to get this project done, so I wasn't thinking about the long term effects. Needless to say, I wasn't caching my MySQL query results. I know, tisk tisk.

He added a new private method to his class (using Zend_Cache to store the date) and dropped the call into his controller to pulled the cached info whenever it needed it.

Brian Moon has announced the release of the latest version of his memproxy tool that uses memcache to create a "server" to save the proxy information.

I put server in quotes because it is really just a PHP script that handles the caching and talking to the application servers. Apache and other HTTP servers already do a good job talking HTTP to a vast myriad of clients. I did not see any reason to reinvent the wheel.

Features of this new version include a TTL for the cached data, minimal dependencies for the application to use and a small code base consisting of two files. It also handles HTTP 1.1 requests, is transparent and applows pages to dynamically be specifically added and removed from the cache.

You can download this latest version from the project's page on the Google Code website.

Brian Shire has posted a new tutorial talking about some of the benefits of caching the functionality of your website (with things like APC, not file or database caching):

Opcode caches save energy, expenses, improve overall user experience on web sites, and it's often one of the simplest optimizations to implement. This article will explain the basics of installing, configuring, and tuning an opcode cache for PHP, the Alternative PHP Cache (APC).

The article focuses on how the APC works and how to get it up and working on your installation (as pulled from the pecl repositories). They look at some of the functions the extension's API includes (like ap_cache_info or ap_store for manual caching) as well as some more advanced topics like locking performance, working with TTL, cache priming and filtering.