The Problem with Future<T> Fetching a product page on an e-commerce site requires at least three data sources: product details, live pricing, and inventory availability. Done sequentially, a 200 ms call to each service adds up to 600 ms. Done in parallel with a properly composed CompletableFuture pipeline, the wall-clock time drops to roughly the slowest of the three — around 200 ms. This article shows exactly how to build that pipeline and everything else you need to write robust async code in Java 8.

Why a New HTTP Client? HttpURLConnection — Java’s HTTP API since Java 1.1 — has deep design problems: Mutable shared state makes it error-prone in multithreaded code No built-in HTTP/2 support No built-in async; non-blocking requires manual thread management Clunky API: setDoOutput(true), getOutputStream(), connect() in sequence No support for reactive streams JEP 321 (Java 11) standardised the HTTP Client API that was incubating since Java 9. The new API lives in java.

Not every operation needs to complete before the response returns. Sending an email, generating a report, publishing an event — these can run in the background. Async processing keeps request latency low while the work continues. @Async — Fire and Forget @SpringBootApplication @EnableAsync public class OrderServiceApplication { } @Service @Slf4j public class NotificationService { @Async // runs in a separate thread public void sendOrderConfirmation(Order order) { log.info("Sending confirmation for order {}", order.