1. References
Use of references are slower than making copies of original data and than manipulate it. So use references only when you have large data structs to save memory. The use of references depends on your system, if your system is CPU bound avoid using the references and if your system is memory bound use references.
2. Persistent Database connections
Generally databases are slower to establish connections so it is better to use persistent db connection. But persistent connections will consume and tie up resources even when not in use so watch your resource limits before making tradeoff.
3. Avoid using regex
Php have good sets of string manipulation functions:
<?php ereg_replace("-","_",$str); ?>Bad
<?php str_replace("-", "_", $str);?>Good
4. mysql_unbuffered_query()
The function mysql_unbuffered_query() is same as mysql_query() except that instead of waiting for the entire query to execute and storing the result in the client API, it makes results available to you as soon as possible and they are not allocated in the client API. So you get access to data quicker and uses less memory. You can't use mysql_num_rows() on the result resource and it is likely to be slightly slower for small selects.
Use of references are slower than making copies of original data and than manipulate it. So use references only when you have large data structs to save memory. The use of references depends on your system, if your system is CPU bound avoid using the references and if your system is memory bound use references.
2. Persistent Database connections
Generally databases are slower to establish connections so it is better to use persistent db connection. But persistent connections will consume and tie up resources even when not in use so watch your resource limits before making tradeoff.
3. Avoid using regex
Php have good sets of string manipulation functions:
<?php ereg_replace("-","_",$str); ?>Bad
<?php str_replace("-", "_", $str);?>Good
4. mysql_unbuffered_query()
The function mysql_unbuffered_query() is same as mysql_query() except that instead of waiting for the entire query to execute and storing the result in the client API, it makes results available to you as soon as possible and they are not allocated in the client API. So you get access to data quicker and uses less memory. You can't use mysql_num_rows() on the result resource and it is likely to be slightly slower for small selects.
