merge and compute

Definition

merge and compute update the value for one key as a single atomic step, so several threads can safely count or update without losing changes. A separate get followed by put is two steps and can lose updates.

Example code

Four threads counting into one map

import java.util.concurrent.ConcurrentHashMap;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.TimeUnit;

class Demo {
    public static void main(String[] args) throws Exception {
        ConcurrentHashMap<String, Integer> counts = new ConcurrentHashMap<>();
        ExecutorService pool = Executors.newFixedThreadPool(4);

        for (int t = 0; t < 4; t++) {
            pool.submit(() -> {
                for (int i = 0; i < 10_000; i++) {
                    counts.merge("visits", 1, Integer::sum);          // atomic read-modify-write
                }
            });
        }
        pool.shutdown();
        pool.awaitTermination(10, TimeUnit.SECONDS);

        System.out.println(counts.get("visits"));                      // always 40000

        counts.compute("visits", (key, old) -> old == null ? 1 : old * 2);   // atomic custom update
        System.out.println(counts.get("visits"));
    }
}

// Output:
// 40000
// 80000