This project contains samples on how to use io.turbodsl on standard applications.
A sample application is implemented using pure Kotlin and TurboDSL:
- Execute 3 jobs (job1, job2, job3) in parallel
- Execute a final job (job4) that will process the results of all previous 3
You can compare:
- Code structure
- Runtime between implementations
- How you can migrate to
TurboDSL
See:
- Pure Kotlin approach
- TurboDSL approach
Output (actual times may vary depending on hardware)
TurboDSL TurboDSL
Kotlin no-logging Diff w/logging Diff
====== ========== ===== ========= =====
Iteration 1 -> 5037ms 5074ms +37ms 5099ms +62ms // First execution always takes more time due to JVM startup
Iteration 2 -> 5004ms 5007ms + 3ms 5007ms + 3ms
Iteration 3 -> 5004ms 5008ms + 4ms 5012ms + 8ms
Iteration 4 -> 5003ms 5008ms + 5ms 5008ms + 5ms
Iteration 5 -> 5005ms 5008ms + 3ms 5005ms 0ms
Iteration 6 -> 5004ms 5006ms + 1ms 5005ms + 1ms
Iteration 7 -> 5006ms 5006ms 0ms 5008ms + 2ms
Iteration 8 -> 5006ms 5006ms 0ms 5005ms - 1ms
Iteration 9 -> 5005ms 5015ms +10ms 5005ms 0ms
Iteration 10 -> 5012ms 5004ms - 8ms 5006ms - 6ms
Roughly, there's a difference of 4ms - without considering the fist execution, even with logging enabled.
This is negligible and the benefits TurboDSL provides outweighs these runtime differences.