]> Google

vasilvv@google.com

tsv QUIC Working Group QUIC namespaces QUIC Stream Namespaces provide an extension to the QUIC protocol that enables multiplexing multiple logical groups of streams within the same connection, while providing flow control isolation. About This Document Status information for this document may be found at . Discussion of this document takes place on the quic Working Group mailing list (), which is archived at . Subscribe at . Source for this draft and an issue tracker can be found at .

Introduction QUIC provides an ordered bytestream abstraction called streams. Streams are subject to various flow control mechanisms that allow a network endpoint to control how much resources a peer is allowed to consume. Some of the flow control mechanisms are scoped to a single stream; others are global to the entire connection. The connection-level flow control mechanisms are a good fit in cases when all of the streams originate from the same entity; however, in cases when multiple logical entities share the same connection, a single global limit may lead to one entity starving another. This document provides a mechanism by which a single QUIC connection can have multiple namespaces, each with its own resource limits for streams.

Conventions and Definitions The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 when, and only when, they appear in all capitals, as shown here.

Namespaces A QUIC namespace is a 62-bit unique ID number. In the initial state, every namespace ID is assumed to exist, but have a MAX_STREAMS number associated with it set to 0 for all types of streams, and a MAX_DATA value of 0 in both directions. A peer opens a namespace by sending a combination of MAX_DATA and MAX_STREAMS frames for that namespace. The recepient may response with either its own MAX_DATA and MAX_STREAMS, confirming the response, or it may close the namespace. Frames that do not have a namespace ID associated with them are said to be a part of the default namespace. Note that there is no way to set a namespace-specific initial_max_stream_data parameters; those remain connection-global.

Frames

NS frame An NS frame (frame type=0x29c5) is a frame that alters the meaning of the frame that comes immediately after it. If the subsequent frame has a stream ID in it, that ID refers to the stream with the corresponding ID in the specified namespace. If the subsequent frame alters connection-global flow control limits, those limits are altered for the namespace in question, instead of the default namespace.

NS Frame Format

The following frames are allowed to follow the NS frame: STREAM, RESET_STREAM, STOP_SENDING, MAX_DATA, MAX_STREAM_DATA, MAX_STREAMS, DATA_BLOCKED, STREAM_DATA_BLOCKED, STREAMS_BLOCKED. Extensions that define their own frames can define their own semantics of interacting with namespaces. If a frame that is not listed above and does not have extension semantics defined for it is prefixed with an NS frame, the recepient MUST close the connection with a PROTOCOL_VIOLATION error code. Same applies to an NS frame that is not followed by anything. Note that this intentionally does not define NS prefix for the DATAGRAM frames , as datagrams already have pre-defined mechanisms for multiplexing (such as ) that may conflict with QUIC stream namespaces, and there is no technical advantage of using an NS frame with datagrams over doing multiplexing within the datagram payload.

CLOSE_NAMESPACE frame A CLOSE_NAMESPACE frame indicates to the peer that the sender will not process any further data received for a given namespace. The sender can discard all of the state related to the namespace after sending this frame.

CLOSE_NAMESPACE Frame Format

Security Considerations TODO Security TODO: discuss the issue where the peer has to remember flow control limits for arbitrary unexpected namespaces.

IANA Considerations TODO: add a transport parameter to negotiate this feature.

References Normative References This document defines the core of the QUIC transport protocol. QUIC provides applications with flow-controlled streams for structured communication, low-latency connection establishment, and network path migration. QUIC includes security measures that ensure confidentiality, integrity, and availability in a range of deployment circumstances. Accompanying documents describe the integration of TLS for key negotiation, loss detection, and an exemplary congestion control algorithm. In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements. RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings. This document defines an extension to the QUIC transport protocol to add support for sending and receiving unreliable datagrams over a QUIC connection. Informative References This document describes HTTP Datagrams, a convention for conveying multiplexed, potentially unreliable datagrams inside an HTTP connection. In HTTP/3, HTTP Datagrams can be sent unreliably using the QUIC DATAGRAM extension. When the QUIC DATAGRAM frame is unavailable or undesirable, HTTP Datagrams can be sent using the Capsule Protocol, which is a more general convention for conveying data in HTTP connections. HTTP Datagrams and the Capsule Protocol are intended for use by HTTP extensions, not applications.

Acknowledgments TODO acknowledge.