Understanding performance bottlenecks

Optimizing Code in Java

Pavlos Kosmetatos

Lead Engineer @Wealthyhood

What are performance bottlenecks?

They are constraints that limit our application's speed and efficiency.

Imagine pouring water through a funnel - the narrowest part determines the flow rate!

Water flowing through funnel

Types of bottlenecks

CPU-bound bottlenecks

Complex calculations
Inefficient algorithms

I/O-bound bottlenecks

Database operations
Network calls
File operations

Memory-bound bottlenecks (covered in later lessons)

Measuring performance with System.nanoTime()

System.nanoTime() - High-precision time measurement
- Measures elapsed time
- Returns nanoseconds (1 billion = 1 second) as a long number
Unlike System.currentTimeMillis(), measures wall clock time

long startTime = System.nanoTime();
// Code to measure
long endTime = System.nanoTime();
long duration = endTime - startTime;

Using System.nanoTime()

// Measuring ArrayList.contains() performance
import java.util.ArrayList;
import java.util.List;

List<Integer> numbers = new ArrayList<>();
for (int i = 0; i < 100000; i++) {
    numbers.add(i);
}


long startTime = System.nanoTime();
boolean found = numbers.contains(99999); // O(n) operation
long endTime = System.nanoTime();

Using System.nanoTime()

long duration = endTime - startTime;
System.out.println("Time taken: " + duration + " ns");
// Convert to milliseconds
System.out.println("Time taken: " + duration / 1_000_000.0 + " ms");

Time taken: 1239000000 ns
Time taken: 1239ms

ArrayList.contains() is an O(n) operation

Best practices for performance measurement

Run multiple measurements

Run multiple measurements

long totalTime = 0;
for (int i = 0; i < 3; i++) {
    long start = System.nanoTime();
    findElementLinear(data, target);
    long iterationTime = System.nanoTime() - start;
    totalTime += iterationTime;
    System.out.println("Run " + (i+1) + ": " + iterationTime + " ns");
}
double averageTime = totalTime / 3.0;
System.out.println("Average: " + (long)averageTime + " ns");

Run 1: 3245678 ns
Run 2: 3198432 ns
Run 3: 3301234 ns
Average: 3248448 ns

Best practices for performance measurement

Run multiple measurements
Compare relative performance

// Assume we have calculated avgTimeA and avgTimeB already
System.out.println("Operation A is" + (avgTimeA / avgTimeB) + "x slower");

Operation A is 4x slower

Let's practice!

Optimizing Code in Java