Measure execution time for a Java method

java

execution-time

profiling

performance

byAlex Kataev·Aug 5, 2024

Efficiently measure a method's runtime speed using System.nanoTime():

long start = System.nanoTime();
methodName(); // Insert joke: Even the snail is faster... Hurry up, methodName!
long duration = System.nanoTime() - start;
System.out.println("Execution time: " + duration + " ns");

This approach gives you the method's runtime with nano-precision accuracy. Speedy, indeed!

Precise timing with NanoTime and Instant

Use System.nanoTime() for precision measurement, while Instant.now() helps when logging requires human-readable time formats.

Instant start = Instant.now();
methodName(); // Insert joke: methodName! Fuel up, it's race time!
Instant end = Instant.now();
Duration duration = Duration.between(start, end);
System.out.println("Execution time: " + duration.toMillis() + " ms");

Pro-tip: For logging purposes, always prefer ISO-8601 time format – because who understands "Fri, 14 Mar, 2022 15:33:00 GMT", right?

Easy timing with Guava's Stopwatch

The utility class Stopwatch, in Google's Guava library, transforms the tiresome task of timing a method's execution into a cakewalk.

Stopwatch stopwatch = Stopwatch.createStarted();
methodName(); // Insert joke: methodName in Stopwatch mode... Let's see who wins!
stopwatch.stop();
System.out.println("Time elapsed: " + stopwatch.elapsed(TimeUnit.MILLISECONDS) + " ms");

Advanced Timing Strategies

Overcoming Random fluctuations with Averaging

Increase accuracy by running and timing the method repeatedly, taking the average time of execution over multiple iterations.

long totalTime = 0;
int iterations = 100;
for (int i = 0; i < iterations; i++) {
    long start = System.nanoTime();
    methodName(); // Insert joke: Again, methodName? Is groundhog day real?
    totalTime += System.nanoTime() - start;
}
long averageTime = totalTime / iterations;
System.out.println("Average execution time: " + averageTime + " ns");

In-depth Analysis with Profilers

NetBeans Profiler or similar tools provide a detailed insight into performance at the method-level, enabling you to precisely spot and eliminate bottlenecks.

Clean Encapsulation with AOP

Aspect-oriented programming (AOP) with Spring's MethodInterceptor segregates timing logic, maintaining the sanity of your code.

Diving Deeper: Overcoming Pitfalls and Exploring Utilities

Averting Pitfalls

Ensure your timing logic doesn't distort performance. Skip heavy CPU or memory-consuming operations within the timing scope and avoid misguiding results.