IPv5

IPv5. Internet Protocol version 5 was an experimental designation assigned to the Internet Stream Protocol (ST), a connection-oriented networking protocol designed for supporting voice, video, and distributed simulation. It was never developed as a successor to the widely deployed IPv4, but its concepts and experimental use informed the development of later protocols. The "version 5" identifier was formally assigned in the late 1970s, creating a gap in the version sequence that was later filled by the definitive next-generation protocol, IPv6.

Overview and History

The history of this protocol is intrinsically linked to early research into real-time multimedia communication. Work began in the 1970s at INWG and was later advanced within the Internet Engineering Task Force (IETF). The protocol was originally known as the Internet Stream Protocol Version 2 (ST-II), an evolution of an earlier concept from researchers like Danny Cohen. Its primary architects, including Jim Forgie of MIT Lincoln Laboratory and others from BBN Technologies, designed it to provide guaranteed quality of service for applications like video conferencing. The Request for Comments process documented its specifications, notably in RFC 1190 and its successor, RFC 1819. This development occurred concurrently with the explosive growth of the ARPANET, highlighting early recognition of the need for services beyond simple datagram delivery.

Technical Specifications

Technically, the protocol operated at the same layer as IPv4 but with a fundamentally different architecture. It established a connection-oriented path, or "stream," between a source and multiple destinations, supporting multicast delivery essential for efficient media distribution. The packet header, defined in the ST specifications, included fields for a unique stream identifier, priority, and a checksum for error detection. Unlike the datagram model of IPv4, it utilized a setup mechanism involving specific control messages (e.g., CONNECT, ACCEPT, DISCONNECT) to manage stream resources along the path through routers. This design required participating routers, such as those running the NSFNET backbone, to maintain per-stream state to handle traffic flow and resource reservations, a concept that would later influence RSVP.

Experimental Use and Deployment

Deployment was limited to controlled experimental networks and specific research projects. One significant testbed was the DARPA-sponsored experimental voice and video network. The U.S. Department of Defense and research institutions like the University of Southern California's Information Sciences Institute utilized it for trials in distributed interactive simulation and battlefield communications. The U.S. Army and U.S. Navy explored its potential for command and control systems. These experiments proved the feasibility of real-time streaming but also revealed complexities in scaling the technology across a global, heterogeneous network like the public Internet.

Relationship to IPv4 and IPv6

The protocol's relationship to IPv4 and IPv6 is one of parallel development rather than succession. It was never intended to replace the addressing or routing framework of IPv4; instead, it was conceived as a complementary protocol that could coexist, carrying specialized traffic. The experimental "version 5" label created a numbering gap, which the IETF explicitly acknowledged when designating the next-generation internet protocol as IPv6. Key lessons from its development, particularly the need for quality of service and efficient multicast, were directly incorporated into the design of IPv6. Features like the flow label field in the IPv6 header and the development of associated protocols like Multicast Listener Discovery reflect this lineage.

Reasons for Non-Adoption

Several critical factors led to its non-adoption as a mainstream standard. The fundamental architectural shift to a connection-oriented model was incompatible with the simple, stateless fabric of the existing Internet, requiring extensive changes to router software from vendors like Cisco Systems and Juniper Networks. The complexity of maintaining stream state in routers presented significant scalability and robustness challenges, especially after experiences with network congestion events like the 1996 Internet collapse. Furthermore, the pressing address exhaustion crisis of IPv4, driven by the growth of networks like the World Wide Web, demanded an immediate solution focused on expanded addressing, which this streaming protocol did not provide. Consequently, the IETF's efforts pivoted decisively toward developing IPv6 as the comprehensive successor.

Category:Internet protocols Category:Experimental networking technologies Category:History of the Internet