dotnet Data Access reader vs adapter, and MySQL APIs vs SQL2k

[sde]

Today I was testing different MySQL APIs for a C# project, and I decided just to make an app.

First I wanted to compare the speed between a data reader and data adapater. Then I wanted to compare those speeds against another MySQL API, and finally I decided why not throw a SQL2k test in there too.

[rdove]

yea...looks like SQLClient blew the others out of the water. And I could have told you the datareader would have been faster. The datareader is readonly and only stores one record at a time in memory, so if you are only displaying records using the datareader is the way to go.

Good test. I'm not surprised that the MYSQL tests came out looking worse, since it is a non-MS product

[sde]

yeah, i knew the reader would be faster, .. just wasn't sure how much faster.

this is even scarier, .. i tested php/mysql this morning too =/ doesn't look like it handles that much data very well. i'm writting an article on it, so i made this table in html and just took a screen shot.

interesting to say the least. it looks like i might get a build of yukon to test with too =) that will be interesting.

this test is only with the reader.

[Belisarius]

If you can, try testing it with Postgresql, as it's suppose to scale better than MySQL. Or even better, Oracle.