What Is Head-of-Line Blocking?

Head-of-line blocking (HOL) is a performance bottleneck where the first item in a queue holds up everything behind it.

When to use/Use Cases

In a network switch, one stuck packet stalls all behind it. In HTTP/1.1, one slow response blocks every request on the same connection.

Example

Think of a one-lane road: if the first car stops, everyone behind must wait.

*Want to go deeper?

Explore Grokking System Design Fundamentals, Grokking the System Design Interview, or Mock Interviews with ex-FAANG engineers to prepare effectively.*

Why Is It Important

HOL blocking reduces throughput and increases latency. For example, in networking, switches may only achieve ~60% of their expected capacity. In web systems, it slows page loads and hurts user experience.

Interview Tips

Explain what HOL blocking is, give a simple analogy, and highlight how newer protocols like HTTP/2 (multiplexing) and HTTP/3 (QUIC) reduce or eliminate it.

Trade-offs

Mitigating HOL blocking often requires parallelism—like multiple queues in switches or multiplexing streams in HTTP/2. This improves performance but increases design complexity.

Pitfalls

Don’t assume it’s “solved.” HTTP/2 fixes it at the application layer, but HOL blocking still exists at the TCP layer. Only HTTP/3 with QUIC eliminates it entirely.