On the IBM developerWorks site today there's a new tutorial showing you how to store shared data directly to a shared memory space of your PHP application.

Once created, and given proper permissions, other processes in the same machine can manipulate those segments by: read, write, and delete. This means that an application written in C can share information with an application written in other languages, such as Java or PHP. They can all share information, as long as they can access and understand that information. [...] This article's proposal is simple, learn how to create and manipulate shared memory segments with PHP and use them to store datasets that other applications can use.

Your PHP installation will need to have been compiled with "enable-shmop" to work with the code in this tutorial. Their examples show how to use the shmop_open, shmop_write and other related functions to read, write, remove and close segments in the shared memory space. They also include an example of using the SimpleSHM library to make it easier to interact with the shared memory space as a standard storage location.

In response to a previous post benchmarking exceptions, Volker Dusch has posted some of his own thoughts and benchmarking results on the same topic.

Some days ago there was a blog post regarding php exception performance in 5.4 and the numbers got reported all over the place. The actually numbers are secondary. The main point is: Don't trust "random" stuff on the Internet when thinking about improving your application performance. You always need to measure things for your self and take care doing so! I've initially trusted the benchmark myself and disgraced the whole post saying: "Well yes, exceptions are slower than if statements but nice that they got faster".

He includes some results with a bit more standardized testing - one run with both 5.3 and 5.4 using XDebug and another with it turned off for both. His results make sense, if you think about them:

So what we learn from that? Running stuff with debugging tools is slower than not doing that. That's why we don't use xDebug in production.

Gonzalo Ayuso has a new post to his blog today looking at the performance of PHP exceptions and how it could effect your application's overall speed.

Sometimes we use exceptions to manage the flow of our scripts. I imagine that the use of exceptions must have a performance lack. Because of that I will perform a small benchmark to test the performance of one simple script throwing exceptions and without them.

His (little) benchmarking scripts are included - both looping 100000 times, one throwing an exception and the other not. The results were pretty obvious - the memory usage was about the same but the speed was about ten times faster without the exceptions (in PHP 5.3). In PHP 5.4, however, the numbers were closer as far as time to run. Obviously, unless you make super heavy use of exceptions, you're not even going to come close to something like this (micro-optimization anyone?).

On PHPMaster.com today there's a new post from Matthew Turland talking about PHP's quest for performance and some of the recent advancements that have made better performing applications even more possible.

While it's sufficient for many users, as PHP sees increased use by large sites like Wikipedia and Facebook, the ability to serve more requests on fewer servers becomes increasingly important. Some efforts have been made in this area in the last few years, both within and outside the PHP internals team. However, understanding exactly what's going on requires a bit of background both in history and concepts.

He goes through some of the origins of the PHP language (from the early days with Rasmus Lerdorf) to the fact that the PHP language itself is interpreted - complete with some of the overhead that comes with that. He also mentions various projects that have tried to compile PHP back down to C to increase performance like Roadsend, HipHop and, most recently, the HipHop virtual machine from Facebook.

In this recent blog post nikic takes an in-depth look at how large PHP arrays really are - how memory is used in the creation and management of these handy PHP variable types.

In this post I want to investigate the memory usage of PHP arrays (and values in general) using the following script as an example, which creates 100000 unique integer array elements and measures the resulting memory usage. [...] How much would you expect it to be? [...] Now try and run the above code. You can do it online if you want. This gives me 14649024 bytes. Yes, you heard right, that's 13.97 MB - eightteen times more than we estimated.

He goes into the details of PHP's memory management and breaks it down into the different totals (for 64 bit and 32 bit OSes) and details on each - zvalue_value, zvalue, cycles collector, Zend MM allocator and the buckets used to isolate one array (hash table/dictionary) from another.

What does this tell us? PHP ain't C. That's all this should tell us. You can't expect that a super dynamic language like PHP has the same highly efficient memory usage that C has. You just can't.

On the VG Tech blog today there's a new post from André Roaldseth about using PHPUnit to test PHP streams, basing the assertions on the data rather than the functionality itself.

Using the memory/temporary stream provided by php:// stream wrapper you can create a stream with read and write access directly to RAM or to a temporary file [using "php://memory"]. This gives you the possibilty to write unit tests that does not rely on a specific file, resource or stream, but rather on data provided by the test itself.

There's no specific code examples here, but you can refer to the stream wrappers section of the PHP manual for more details on this and other handy built-in streams. Once created, it can then be used just as any other stream resource can. This could be useful to provide mocks in your testing, replacing any other stream-able resource with a "memory" or "temp" placeholder.

Matt Farina has a new post today looking at an "underutilized gem" he's found in the offerings of the Standard PHP Library (SPL) - the SplFixedArray.

Arrays in PHP are not arrays per the typical array data type. Instead, as Matt Butcher recently pointed out arrays in PHP are similar to hashes in other languages. This can be a very important point to know when tracking down bugs in code and to programmers coming to PHP from other languages. But, what if we wanted something like a traditional array data type? Maybe something that preserved numeric order. Enter SplFixedArray.

He gives an example of using the SplFixedArray object versus the normal array variables in a simple PHP snippet showing the preservation of numbering order. He also touches on the memory consumption difference between the two, with the fixed array coming in quite a bit lower than the normal array data type (around 25% based on his basic testing). There are some catches to using it, though including incompatibility with array methods and the fact that it doesn't implement things like Iterator or Countable interfaces.

While admitting that micro-optimizations aren't usually worth the time that's taken to worry about them, Gonzalo Ayuso has thrown together some array read benchmarks to show the difference, if any, in where array values are fetched.

Normally our code is coded once and executed thousands of times, and we must keep in mind that CPU time many times means money. We need to balance it (as always). But, WTH. I like micro-optimizations, and here comes another one: Checking the access to arrays with PHP.

He sets up three different options and tests the memory consumption and run time for each:

• Referencing a value from a large array outside a for loop
• Referencing a value from a large array inside a for loop
• Echoing out the value from a large array inside a for loop

Not surprisingly, all three approaches yield just about the same results. It probably has more to do with the size of the large array than how it's accessed. The fetch outside the for loop did come in slightly under the others, but not enough to worry about it.

In the process of some development he's been doing on various shared modules for PHP, Derick Rethans stumbled across an issue with using Valgrind to test his code:

While testing whether I correctly free all memory with Valgrind, I ran into the issue where I couldn't see the stack frames of where the memory leaks occurred in the extensions, and once I even ran into a Valgrind bug. The reason why Valgrind could not show the function names belonging to the stack frames is because PHP had already unloaded the shared extensions from memory.

A work-around he found was compiling the modules, but he wanted something "more correct" and less of a hassle. As a result he added a check for the ZEND_DONT_UNLOAD_MODULES environment flag to the PHP core to handles this case specifically. He includes a snippet of example code showing the Valgrind results with and without the flag.

Shay Ben Moshe has put together a quick post today where he benchmarks array handling performance differences between PHP's native array and the newer SplFixedArray data structure that's a part of the Standard PHP Library that comes with any recent version of the language.

In PHP, arrays are one of the most fundamental data structures. We use them everywhere. They are very flexible, because they are implemented as associative arrays, and therefore let us use both string and integer keys. They are also unlimited in size, in most languages arrays are fixed-sized, but this is not the case in PHP. With that in mind, there still is a drawback. It damages performance. The solution for this problem may be SplFixedArray. But, it is not a perfect solution.

He points out two major differences - the SplFixedArray is, well, a fixed size and the fact that it can only use integer keys (no associative arrays here). He created three tests to compare the performance of the two:

• Writing data to the array
• Reading data from the array
• Getting a random value from the array

Each of these are measured in terms of runtime and/or memory usage. If you'd like to try out the tests for yourself, you can download the files needed. I won't cover the results of the tests here, though - you'll need to visit the post for that!