stream() vs parallelStream()

Java 8 introduced Streams to process collections in a functional style. Streams can operate in:

• Sequential mode → stream()
• Parallel mode → parallelStream()

Understanding the difference is critical for writing performant and safe code.

---

Basic Definition

stream()	parallelStream()
Sequential processing	Parallel processing
Single thread	Multiple threads
Order preserved	Order not guaranteed
Predictable behavior	Non-deterministic execution

---

How stream() Works

List<Integer> list = List.of(1, 2, 3, 4);

list.stream().forEach(System.out::println);

Execution:

1 → 2 → 3 → 4

• Uses main thread
• Processes one element at a time
• Maintains encounter order

---

How parallelStream() Works

list.parallelStream().forEach(System.out::println);

Possible execution:

3 → 1 → 4 → 2

Why?

• Uses ForkJoinPool.commonPool()
• Splits data using Spliterator
• Processes chunks in parallel threads
• Merges results

---

Internal Working

stream()

Main Thread
↓
Sequential processing

parallelStream()

ForkJoinPool.commonPool()
↓
Split data
↓
Process in worker threads
↓
Combine results

Default thread count:

Number of CPU cores - 1

---

Aspect	stream()	parallelStream()
Thread Model	Single thread	Multiple threads
Order	Preserved	Not guaranteed
Performance	Stable	Depends on workload
Safety	Safer	Risk of concurrency issues
Best For	Small/medium data	Large CPU-bound tasks

---