Facebook gave a MySQL Tech Talk where they talked about many things MySQL, but one of the more subtle and interesting points was their focus on c

Facebook at 13 Million Queries Per Second Recommends: Minimize Request Variance - High Scalability -

submited by
Style Pass
2021-07-27 12:00:06

Facebook gave a MySQL Tech Talk where they talked about many things MySQL, but one of the more subtle and interesting points was their focus on controlling the variance of request response times and not just worrying about maximizing queries per second.

Excellent information. All I ever read is how to get the highest queries/s. This approach made me turn my head a bit at first, but then it made complete sense.

Anyone know why Facebook chose mysql over postgresql? I'm not suggesting that I would do either and I'm sure there will be plenty of people with their own personal opinions of which is best but I'm truly curious for Facebook's reasons.

Queries/sec is an important measure that should not be ignored. These numbers are the only reasonable way of benchmarking and comparing different architectures. When you're designing your system, those numbers, however abstract and meaningless, are about the only way of giving you an idea of the relative merits of each choice.

But when you go live and measure your actual performance, what you should get is essentially a nice, stable performance baseline under average load. If only the real world was so nice. Instead, you probably sometimes get your expected performance, and have peaks and valleys where things do not quite go as expected.

Leave a Comment
Related Posts