Some Stream classes use very fine-grained locking.

In particular, writeInt is defined like this:

        out.write((v >>> 24) & 0xFF);
        out.write((v >>> 16) & 0xFF);
        out.write((v >>> 8) & 0xFF);
        out.write((v >>> 0) & 0xFF);
        incCount(4);

Unfortunately, ByteArrayOutputStream.write(byte) is defined like this:

    public synchronized void write(int b) {
        ensureCapacity(count + 1);
        buf[count] = (byte) b;
        count += 1;
    }

so we acquire and release a lock for every byte that is output.

With biased locking we don't see the pain because the acquire and release operations are basically no-ops, and C2 optimizes the whole lot away. Without biased locking, things are very much worse.

For example, writing 4kb of ints goes from 17.3 us/op to 53.9 us/op when biased locking is disabled:

+UseBiasedLocking DataOutputStreamTest.dataOutputStreamOverByteArray avgt 6 53.895 ± 5.126 us/op
-UseBiasedLocking DataOutputStreamTest.dataOutputStreamOverByteArray avgt 6 17.291 ± 4.430 us/op

There are refactorings of DataOutputStream we can do to mitigate this.