Performance‑Optimized Alternatives to Common PHP Functions

Usually I write straightforward PHP code using built‑in functions, but for some tasks I have discovered significantly faster alternatives. All tests were performed on a local web server running PHP 7.4.

1. Remove Duplicates

Goal: deduplicate a large one‑dimensional array.

Typical approach: array_unique($array); Faster alternative: array_keys(array_flip($array)); Benchmark (4 million elements, 3 million duplicates):

Method

Execution time

array_unique

787.31 ms

array_keys array_flip

434.03 ms

The alternative is about 1.8× faster (≈45 % improvement) and works for one‑dimensional arrays because array_flip swaps keys and values.

2. Get a Random Array Element

Goal: retrieve a random value from a large array.

Typical approach: array_rand($array); Faster alternative: $array[mt_rand(0, count($array) - 1)]; Benchmark (5 million elements):

Method

Execution time

array_rand

25.99 µs

mt_rand

0.95 µs

The alternative is roughly 27× faster (≈96 % improvement) and, on average, 8× faster despite both using the same underlying random algorithm since PHP 7.1.

3. Alphanumeric Character Test

Goal: verify that a string contains only letters and digits.

Typical approach: preg_match('/[a-zA-Z0-9]+/', $string); Faster alternative: ctype_alnum($string); Benchmark (100 k+ test strings):

Method

Time

preg_match

1.539 ms

ctype_alnum

2.06 ms

The ctype_alnum version is about 7.5× faster (≈87 % improvement) and on average 4× faster. The same principle applies to ctype_alpha and ctype_digit.

4. Replace Substring

Goal: replace part of a string with another substring.

Typical approach: str_replace('a', 'b', $string); Faster alternative: strtr($string, 'a', 'b'); Benchmark (5 million random strings):

Method

Time

str_replace

6.76 ms

strtr

305.59 ms

Here strtr is about 2.2× faster (≈55 % improvement) on average.

Additional Performance Tips

Prefer JSON over XML.

Declare variables before loops instead of inside each iteration.

Avoid function calls in loop headers (e.g., for ($i = 0; $i < count($array); $i++)).

Unset variables that consume memory when they are no longer needed.

Prefer switch statements over multiple if checks.

Use require/include instead of require_once/include_once when possible to improve opcode caching.

While premature optimization is often debated, real‑world production bottlenecks usually lie elsewhere (e.g., database queries), but when a faster, maintainable alternative exists—especially for regex‑heavy code—it is worth adopting.

Conclusion

Even with PHP 7.4’s inherent speed gains, alternative functions can still deliver noticeable performance improvements across common array and string operations.