Databases Reference
In-Depth Information
[3] sc:22 lt:0 rt:0 fl:0
[4] sc:22 lt:0 rt:0 fl:0
in 30 sec.
<Raw Results2(sum ver.)>
[0] sc:221 lt:0 rt:0 fl:0
[1] sc:221 lt:0 rt:0 fl:0
[2] sc:23 lt:0 rt:0 fl:0
[3] sc:22 lt:0 rt:0 fl:0
[4] sc:22 lt:0 rt:0 fl:0
<Constraint Check> (all must be [OK])
[transaction percentage]
Payment: 43.42% (>=43.0%) [OK]
Order-Status: 4.52% (>= 4.0%) [OK]
Delivery: 4.32% (>= 4.0%) [OK]
Stock-Level: 4.32% (>= 4.0%) [OK]
[response time (at least 90% passed)]
New-Order: 100.00% [OK]
Payment: 100.00% [OK]
Order-Status: 100.00% [OK]
Delivery: 100.00% [OK]
Stock-Level: 100.00% [OK]
<TpmC>
442.000 TpmC
The very last line is the benchmark result: the number of transactions per minute that
the benchmark achieved.
10
If you see aberrant results in the lines immediately preceding
this, such as the constraint check lines, you can examine the response-time histograms
and other verbose output for clues about what was going wrong. Of course, you should
have used scripts such as those we showed earlier in this chapter as well, so you should
also have detailed diagnostic and performance data about what the server was doing
during the benchmark run.
Summary
Everyone who uses MySQL has reasons to learn the basics of benchmarking it. Bench-
marking is not just a practical activity for solving business problems, it's also highly
educational. Learning how to frame a problem in such a way that a benchmark can
help provide an answer is analogous to working from word problems to setting up
equations in a math course. Phrasing the question correctly, choosing the right bench-
mark to answer the question, selecting the benchmark duration and parameters, run-
ning the benchmark, collecting the data, and analyzing the results will all make you a
much better MySQL user.
10. We ran this benchmark on a laptop for demonstration purposes only. Real servers should perform much
faster.
