Classless Inter-Domain Routing

Classless Inter-Domain Routing
Name	Classless Inter-Domain Routing
Caption	CIDR block aggregation example
Introduced	1993
Designer	Internet Engineering Task Force
Purpose	IPv4 address allocation and routing aggregation

Classless Inter-Domain Routing Classless Inter-Domain Routing was introduced to extend IPv4 address allocation and improve routing efficiency, reducing Internet routing table growth and enabling more flexible subnetting. The proposal emerged from engineering work within the Internet Engineering Task Force and was formalized through documents influenced by operators at BBN Technologies, Cisco Systems, and researchers at Stanford University and MIT.

Background and Rationale

CIDR originated as a response to address exhaustion and routing table expansion observed by operators at MERIT Network, ARPA, National Science Foundation, and commercial providers such as AT&T and Sprint. Debates at meetings of the Internet Assigned Numbers Authority and working groups of the IETF reflected input from implementers at Juniper Networks, 3Com, and Bell Labs. Historical events like the rapid commercial growth in the 1980s and early 1990s, including policy shifts at UUNET and deployment by backbone providers such as MAE-East, motivated technical reforms led by authors from Cisco Systems and academics affiliated with UC Berkeley.

CIDR Notation and Addressing

CIDR introduced a compact representation for IPv4 prefixes using a slash notation inspired by subnetting practices used at Xerox PARC and research projects at Carnegie Mellon University. The notation expresses network prefixes as an address followed by a prefix length, aligning operations on routers from vendors such as Cisco Systems, Juniper Networks, and Nokia and software stacks in FreeBSD, Linux, and Windows NT. Implementations referenced standards maintained by the IETF and the allocation frameworks used by regional registries like ARIN, RIPE NCC, and APNIC.

Routing and Aggregation

CIDR enabled route aggregation techniques deployed on backbone networks operated by Verizon Business, Level 3 Communications, and NTT Communications, allowing multiple contiguous prefixes to be represented as a single route. Aggregation reduced entries in global tables handled by route reflectors and BGP implementations produced by Cisco Systems, Juniper Networks, and open-source projects like Quagga and BIRD. Operational experiences from exchange points including LINX, AMS-IX, and DE-CIX informed best practices for prefix announcement and summary routes used by network operators at companies such as Google, Amazon Web Services, and Microsoft.

Implementation and Deployment

Deployment required updates to Border Gateway Protocol implementations and configuration practices used by operators at Sprint, AT&T, Verizon Business, and academic networks such as CERNET and Internet2. Router vendors including Cisco Systems, Juniper Networks, Alcatel-Lucent, and Huawei integrated CIDR-aware route selection and longest-prefix-match algorithms derived from academic work at Princeton University and University of California, Los Angeles. Regional Internet Registries like LACNIC, AFRINIC, and ARIN adjusted allocation policies affecting organizations including IBM, HP, and Oracle.

Impact on Internet Architecture and Scalability

CIDR substantially slowed the growth of the global BGP routing table faced by operators at major backbones such as Level 3 Communications and content providers such as Akamai Technologies and Cloudflare. The technique affected address policy decisions made by IANA and influenced transition planning by vendors including Cisco Systems and service providers such as Comcast and Deutsche Telekom. CIDR's role in enabling efficient use of IPv4 address space was linked to subsequent discussions about IPv6 adoption promoted by organizations including IETF, IEEE, and ISOC.

Limitations and Criticisms

Critics from academic centers like MIT and Stanford University and network operators at entities such as Small ISP Association raised concerns about limits to aggregation when multihoming customers require de-aggregation, leading to larger routing tables in practice. Operational constraints highlighted by exchanges like AMS-IX and policy debates at RIPE NCC showed tension between aggregation goals and business requirements of enterprises like Amazon Web Services and Google. Further, the finite nature of IPv4 addressed by registries including APNIC and ARIN ultimately shifted emphasis toward IPv6 initiatives advocated by IETF and ISOC.

Category:Internet protocols