On October 10th, 2007 Hubert Roksor (not verified) says:
Hi,
I've found this article via the PostgreSQL's site. I am more familiar with MySQL than PostgreSQL and I'm not trying to say that either one is faster or more performant than the other. However, I'm very surprised by the results of your benchmark and I suspect that your MySQL wasn't optimally configured.
I read that you used AB, that means you retrieved the same page repeatly, therefore if the query cache was working, only the first page would have actually executed the queries and all subsequent pages' queries' results would have been pulled from the cache directly. That's an ideal situation for MySQL's query cache and should have tremendously biased the benchmark towards MySQL, that's why I'm very surprised to see PostgreSQL executing twice as many queries (pages) as MySQL in your second benchmark. If you decide to run a new batch of tests, try comparing the stats (SHOW STATUS) before and after running ab, and see if the cache was used.
Misconfiguration?
Hi,
I've found this article via the PostgreSQL's site. I am more familiar with MySQL than PostgreSQL and I'm not trying to say that either one is faster or more performant than the other. However, I'm very surprised by the results of your benchmark and I suspect that your MySQL wasn't optimally configured.
I read that you used AB, that means you retrieved the same page repeatly, therefore if the query cache was working, only the first page would have actually executed the queries and all subsequent pages' queries' results would have been pulled from the cache directly. That's an ideal situation for MySQL's query cache and should have tremendously biased the benchmark towards MySQL, that's why I'm very surprised to see PostgreSQL executing twice as many queries (pages) as MySQL in your second benchmark. If you decide to run a new batch of tests, try comparing the stats (SHOW STATUS) before and after running ab, and see if the cache was used.
Regards,
HR