Head-of-line blocking

In computer networking, head-of-line blocking (HOL blocking) refers to a performance bottleneck that occurs when a queue of packets is held up by the first packet in the queue, even though other packets in the queue could be processed.

In HTTP/1.1, HOL blocking occurs when a client sends multiple requests to a server over a particular TCP connection, and a response on that connection is delayed for any reason — such as network congestion, TCP slow start, or, problems in transit. HTTP/1.1 requests are sent sequentially per TCP connection, so a delay in receiving a response blocks the next request-response exchange.

A mechanism called HTTP pipelining tried to work around this, where multiple requests were sent off by a client without waiting for any responses. Pipelining proved tricky to implement in reality, so this mechanism is rarely used, if ever, and most browsers no longer support it.

HTTP/2 addresses the HOL blocking problems in HTTP/1.1 through request multiplexing. Multiplexing allows a single TCP connection to interleave requests and responses in numbered streams. A client can send many requests over a single connection without waiting for earlier responses. Note that while HOL blocking has been fixed in HTTP/2, it is still a problem at the transport (TCP) layer level.