Why QUIC Is Revolutionizing HTTP: From HTTP/1.1 to HTTP/3 Explained

HTTP History

1991 – HTTP/1.1 was standardized. 2009 – Google designed SPDY over TCP. 2013 – QUIC was introduced. 2015 – HTTP/2 was standardized. 2018 – HTTP/3 (HTTP over QUIC) was standardized.

HTTP/3 uses UDP to retain QUIC's speed while preserving TLS security.

QUIC Protocol Overview

QUIC (Quick UDP Internet Connections) is a UDP‑based protocol that combines the advantages of TCP, TLS, and HTTP/2 while optimizing performance.

QUIC replaces TCP/SSL/TLS at the transport layer. Any application‑layer protocol (HTTP, FTP, IMAP, etc.) can run over QUIC; when HTTP runs over QUIC it is called HTTP/3.

RTT Connection Establishment

HTTP/2 requires 3 RTT (or 2 RTT with session reuse, 1 RTT with TLS 1.3). HTTP/3 establishes a connection with only 1 RTT for the first handshake and 0 RTT for subsequent connections.

Client sends an Inchoate Client Hello.

Server replies with a Rejection containing its DH parameters (g, p, A).

Client generates its own DH parameters, computes the shared secret K, encrypts HTTP data with K, and sends B and the encrypted data.

Server derives the same secret, decrypts the data, updates its DH values, and sends a Server Hello with the new public key and HTTP response.

Connection Migration

TCP connections are bound to a four‑tuple (source IP, source port, destination IP, destination port); any change forces a new connection. QUIC uses a 64‑bit Connection ID, so the connection persists across IP or port changes.

Head‑of‑Line Blocking & Multiplexing

Both HTTP/1.1 and HTTP/2 suffer from head‑of‑line (HoL) blocking at the TCP level. HTTP/2 mitigates HoL at the request level by multiplexing streams over a single TCP connection, but TCP’s own HoL remains.

Because QUIC operates over UDP, loss of a single packet does not stall the entire connection; only the missing packet is retransmitted.

Congestion Control

QUIC adopts TCP’s congestion‑control concepts (slow start, congestion avoidance, fast retransmit, fast recovery) and implements the Cubic algorithm with additional features:

Hot‑plugging: congestion‑control algorithms can be changed at the application level without kernel modifications.

Forward Error Correction (FEC): data is split into 10 packets, an XOR‑based FEC packet is added, allowing recovery of a single lost packet.

Monotonically increasing Packet Number: eliminates ambiguity between original and retransmitted packets.

ACK Delay: accounts for receiver processing time, yielding more accurate RTT measurements.

More ACK blocks: QUIC can carry up to 256 ACK blocks, reducing unnecessary retransmissions in lossy networks.

Flow Control

TCP performs flow control only at the connection level using a sliding window. QUIC introduces two levels of flow control:

Stream‑level: each stream has its own receive window that shrinks as data arrives and expands when the receiver processes data.

Connection‑level: the overall receive window is the sum of all stream windows, allowing coordinated throttling across streams.

When the unconsumed portion of a stream falls below half of its maximum window, the receiver sends a WINDOW_UPDATE frame, allowing the sender to transmit more data.