-
Notifications
You must be signed in to change notification settings - Fork 544
Overhead
LinShunKang edited this page Sep 1, 2019
·
1 revision
MyPerf4J's overhead is so minimal that it is difficult to measure on most applications.
We have to use super fast method with response times in the nanoseconds in order to reliably detect and measure MyPerf4J's overhead.
The benchmark below is run against a not inline empty method (benchmarking is performed using the JMH).
- JMH version: 1.21
- VM version: JDK 1.7.0_80, Java HotSpot(TM) 64-Bit Server VM, 24.80-b11
- OS: macOS Mojave 10.14.6
- CPU: Intel(R) Core(TM) i9-8950HK CPU @ 2.90GHz
Response times | Baseline | with MyPerf4J | Overhead |
---|---|---|---|
Average | 2.6 ns | 56.8 ns | 54.2 ns |
90th percentile | ≈ 0 ns | ≈ 0 ns | ≈ 0 ns |
95th percentile | ≈ 0 ns | 1000.0 ns | 1000.0 ns |
99th percentile | 1000.0 ns | 1000.0 ns | 0.0 ns |
99.9th percentile | 1000.0 ns | 1000.0 ns | 0.0 ns |
Benchmark details can be found at MyPerf4J-Benchmark
* Home
- Chinese-Doc
-
English Doc
- MyPerf4J
- Time Series Database
- Log Collector
- Visualization Platform