Re: GetRows equivalent?

First off, we're not using MySQL. However, just to test this theory, I've written a couple small benchmarks. I created a 500 element array, and made a 500 line MySQL table out of this. Also, I took that 500 element array, serialized it, and stored it in another table as a blob. Selecting the columns from the first table, looping thru line by line (fetch_row, fetch_array, whatever), and echo the results took .053 seconds. Selecting the blob, unserializing it, and stepping thru that array came up as .026 seconds. That's a big difference - looks like 100% faster doing it the 'single grab' way. I don't know enough about the innards of how the specific drivers work to know if they indeed pull everything back, or piece it together step by step. There does in fact seem to be some overhead calling the fetch_array 500 times. Whether that's simply the overhead of calling a function or not, I don't know. Perhaps someone could shed a bit more light on how the internals of PHP/Zend work in this respect. I realize .03 seconds doesn't mean much to some people, but it's the percentage difference that interests us. We're trying to optimize things as much as possible. Publishing certain 'semi-static' data to a serialized structure rather than pulling it back line by line seems to be faster, and if we've got 300 people simultaneously banging on this, .03 seconds can add up. The only harm I can see is memory usage - pulling large result sets back for multiple people may eat up RAM a lot quicker than single file. Anyone have any thoughts/experience on this? Soeren Staun-Pedersen wrote: > > Is there anything in PHP db drivers that could be considered > > the equivalent of GetRows from the ASP/ODBC world? In a nutshell, > > rather than looping thru a result set one row at a time, you > > can 'GetRows' and the entire result set will be returned as an > > array that you can loop thru locally, rather than continually > > pulling from the DB driver. > > As far as I know there isn't. I don't know if it is considered harmful, > but I would. The result set you get from a, e.g. mysql_query(), call is > gathered in one fell swoop. It is not gathered bit by bit from the server > everytime you read a row from it. > > Doing what you suggest would mean moving one large datastructure into > another large datastructure.. that cannot be a speed improvement. > > You can go an awful long way with this little construct: > > $query = mysql_query("some sql"); > > while($row = mysql_fetch_array($query) > { > echo $row["some_column"]; > echo $row["some_other_column"]; > } > > Cheers, > > Soeren Staun-Pedersen - staun@netwarriors.org > ------ > "The internet is full, beat it" > - Me. > > -- > PHP General Mailing List (http://www.php.net/) > To unsubscribe, e-mail: php-general-unsubscribe@lists.php.net > For additional commands, e-mail: php-general-help@lists.php.net > To contact the list administrators, e-mail: php-list-admin@lists.php.net -- ========================== Michael Kimsal http://www.tapinternet.com 734-480-9961