RFC 1678

IPng Requirements of Large Corporate Networks

📅 published August 1994, from ietf

IPng Requirements of Large Corporate Networks 🎀

okay bestie, this is RFC 1678 💼✨ it's filed under the Network Working Group and written by E. Britton and J. Tavs from IBM. It's an Informational doc dated August 1994, so it's not a standard, just a super smart wishlist for the Internet community 🦋. The title block doesn't list anything it obsoletes or updates, so it's doing its own thing 💅

Status of this Memo 📜

this memo gives info to the Internet community 🌸 it doesn't specify an Internet standard of any kind, and distribution is unlimited, so anyone can pass it around 💌

Abstract 🍓

this doc was submitted to the IETF IPng area in response to RFC 1550. publishing it doesn't mean the IPng area signed off on any of the ideas inside 🙈 comments should go to the big-internet@munnari.oz.au mailing list. it's a draft that sums up some of the requirements of large corporate networks for the next generation of the Internet protocol suite 🧚

Executive Overview 👑

okay so more and more corporations are using TCP/IP for their mission-critical applications 😭 and they're bringing extra requirements (listed below) that would make TCP/IP even more appealing to businesses. these are requirements, not solutions, so we include capabilities that might be provided in protocol layers other than the one IPv4 occupies. i.e., these items might sit outside the scope usually imagined for IPng, but we'll still call them IPng requirements anyway ✨ when we mention potential solutions, it's not because they're the best approach, it's just to make the requirement clearer 💅

among business users, the major requirements for IPng are:

smooth migration from, and coexistence with, IPv4;
predictable levels of service for predictable costs;
security; and
accommodation of multiple protocol suites.

we also mention several more specific requirements 🎀

IPng must have a viable strategy for migration from, and coexistence with, IPv4. IPv4 and IPng have to coexist well, because they'll need to for several years 🦋 to get IPv4 users to upgrade, IPng must offer compelling advantages and an easy migration path 🔥

corporate networks must meet promised levels of service while controlling costs through efficient use of resources 💰 the IETF should consider both technical solutions (like service classes and priorities) and administrative ones (like accounting) to promote economy 🌈

many businesses won't connect to a network until they're confident it won't significantly threaten the confidentiality, integrity, or availability of their data 🔒

corporations tend to use multiple protocols 🧚 lots of forces stop them from settling on just one, like diverse installed bases, skills, technical factors, and the general trend toward corporate decentralization. the IETF needs a strategy for heterogeneity that's flexible enough to accommodate the principal multiprotocol techniques, including multiprotocol transport, tunneling, and link sharing 💜

some of these requirements might be met by deploying existing Internet architectures more widely (e.g., Generic Security Service and IPv4 type of service) ✨ the current Internet protocols could be enhanced to cover most of the remaining requirements of commercial users while keeping IPv4. still, some corporations will be scared off TCP/IP by the publicity about the address space until the IETF sets a direction for its expansion 😬

Migration and Coexistence 🦋

as IPv4 use keeps growing, the day may come when no IPv4 network addresses are left, and no new networks can connect to the Internet 💀 Classless Inter-Domain Routing (CIDR, RFC 1519) and careful gleaning of the address space will push that cutoff back for several years. the hundreds of millions of people on networks that do get IPv4 addresses won't be hit directly by the address space running out, but they'll miss out on talking to the people who were less lucky 🥺

because the Internet is too big for all its users to cut over to IPng quickly, IPng must coexist well with IPv4 🫶 plus, IPv4 users won't upgrade without a compelling reason. access to new services won't be a strong motivator, since new services will want to support both IPng and IPv4 users. only services that can't exist on IPv4 will be willing to use IPng exclusively 😤 and if IPng needs more resources (like storage, memory, or administrative complexity) than IPv4, users won't install it unless it has clear benefits over IPv4 📉 the millions of users on low-end systems (DOS, sub-notebooks) might never be able to use IPng if it takes more memory 💔

so there's going to be a long period of coexistence between IPng and IPv4, and the coexistence needs to be quite painless, not based on any assumption that IPv4 use will drop off quickly 🌸

Service Level Agreements 💅

if a corporation depends on its network for apps that are critical to the business (like airlines for reservations, and brokerages for stock and bond trades), then the corporation insists the network deliver the needed service level for a predictable cost 😌 that way they can plan for it in their budget ahead of time 💖 a service level agreement (SLA) is a contract between the network's provider and its users. it defines the service level a user will see and the cost tied to that level of service. SLA measurements may include response times (average and maximum), availability percentages, number of active sessions, throughput rates, etc. businesses need to predict and guarantee service levels and costs (routing capacity, link bandwidth, etc.) for their traffic patterns on a TCP/IP network 🧾

IPng should allow control of the cost of networking, which is a major concern for corporations 💸 teleprocessing lines are a big cost in corporate networks. the cost per bit-per-second tends to be lower on higher-bandwidth links, but high-bandwidth links can be hard to get, especially in emerging nations. in lots of places it's hard to get a 64 kpbs line, and T1 service might not exist at all 😭 lead times can also be over six months. even in the US, transcontinental T1 service is pricey enough to push people toward high utilization 📈 cost-conscious businesses want IPng to allow high utilization of teleprocessing links, but without needing excessive processing power to get there ⚡

there's been a lot of speculation about goodput through congested routes when using the Internet's current congestion control algorithms. instead, it should be measured across a range of realistic cases 🔍 if peak-busy-hour goodput under congestion is near the theoretical maximum, publicize the data and move on to other requirements 🎉 if not, the IETF should seek a better standard (e.g., they might explore XTP's adaptive rate-based approach and other proposals) 🛠️

functions like class of service and priority, which let an enterprise control its use of bandwidth, may also help meet service level agreements ⚠️

on one hand, people say that the absence of these functions holds back TCP/IP usage in corporate networks, especially when predictable interactive response times are needed 🤔 on the other hand, few vendors have felt motivated to implement TCP's architected type-of-service, and priority tends to be handled in a non-standard way (e.g., making sure interactive well-known ports, like Telnet, get faster response times than non-interactive well-known ports, like file transfer) 🙃 the IETF should sort out these apparently conflicting views. if the ad hoc techniques can be shown to be adequate, then they should be standardized 🫶 otherwise, effective techniques should be developed and standardized 🧚

commercial users often want a higher level of service for a higher cost, or a lower level of service for a lower cost. e.g., some businesses pay top dollar to make sure response times are fast during business hours, but choose cheaper satellite services for data backup at night 🌙 widespread use of IPv4's type-of-service markings might satisfy this requirement 🎀

to discourage waste of bandwidth and other expensive resources, corporations want to account for their use 🧾 direct cost recovery would let an entity measure and benchmark its efficiency with minimal economic distortion ✨ alternatives, like putting these costs into corporate overhead or charging per connection, make sense when the administrative cost of usage-based accounting is high enough to cause more economic distortion than the alternatives would 🌀 for example, connection-based costs alone may be fine for a resource (like LAN bandwidth) that isn't scarce or expensive, but a mix of a connection cost and a usage cost may be a better fit for a scarcer or pricier resource (like WAN bandwidth) 💜 balance must be kept between the overhead of accounting and the granularity of cost allocation ⚖️

Security 🔐

many corporations will stick with their private networks until public ones can guarantee equivalent confidentiality, integrity, and availability 🙅‍♀️ it's not clear that additional architecture is needed to meet this requirement. maybe more widespread use of existing security technology would be enough 🤷‍♀️ for example, the Internet could encourage wide deployment of Generic Security Service, and then ask for feedback on whether additional security requirements need to be met 🔑 also, businesses care so much about network cost control mechanisms that they want them secured against tampering 🚨 IPng should not interfere with firewalls, which many corporations consider essential 🧱

Heterogeneity 🌈

corporate users want the Internet to accommodate multiple protocol suites 🧩 several different suites are growing in use, and some older ones will be around for many more years. lots of people wish there were only one protocol in the world, but there's little agreement on which one it should be 😅

since the marketplace hasn't settled on one way to handle multiple protocols, IPng should be flexible enough to accommodate a variety of technical approaches to heterogeneity 🦋 for example, most networking protocols assume they'll be the dominant protocol that transports all the others. protocol designers should pay more attention to making their protocols easy for others to transport 🫶 IPng needs to be flexible enough to handle the major multiprotocol trends, including multiprotocol transport networking (for an example, see X/OPEN document G306), tunneling (both IP being the tunnel and being tunneled), and link sharing (e.g., point-to-point protocol and frame relay) 💖 fair sharing of bandwidth by protocols with different congestion control mechanisms is a particularly interesting subject 👀

Flow and Resource Reservation ✨

Okay so corporate users are getting obsessed with sending non-isochronous and isochronous info together over the same link 🎀. IPng should coexist effectively with the isochronous protocols being built for the Internet 💖.

The Internet protocols should totally take advantage of whatever an underlying fast packet switching service offers 🚀. Constant-bit-rate and variable-bit-rate services usually need traffic descriptors to be specified and conformed to, plus quality-of-service objectives that come from applications or users 📋. The Internet's isochronous protocols should have ways to make use of the multimedia services that fast packet switching networks will offer, and they must ensure that quality-of-service guarantees stay preserved all the way up the protocol stacks to the applications 🫶.

Protocols riding on available-bit-rate services can get better bandwidth utilization if they react to congestion messages from a fast packet switching network, and if they think about what happens when cells get discarded 🌊. (Ex: if one cell of an IP datagram gets discarded, it's a total waste to keep forwarding the rest of the cells in that datagram 😭. Also, selective retransmit should be revisited in this context 🔁.)

When the Internet protocol suite lets non-isochronous and isochronous traffic mix on one medium, it should give us ways to discourage inappropriate reservation of resources 🙅‍♀️. Like, a Telnet connection probably doesn't need to reserve 45Mbps, bestie 💀. Accounting, class-of-service, and well-known-port distinctions are possible ways to meet that requirement ✨.

Mobile Hosts 📱

Wireless tech opens up brand-new TCP/IP apps made specifically for mobile hosts 🌸. Besides working with the organizations building wireless standards, the IETF also should encourage the specification of new TCP/IP applications enabled by wireless, like connectionless messaging 💬.

IPng should deal well with the characteristics (delay, error rates4, etc.) peculiar to wireless 📡.

Topological flexibility 🦋

Today a TCP/IP host that moves to a different subnet needs a new IP address 😤. Those moves and changes can turn into a significant administrative cost 💸. Plus, mobile hosts need flexible topology, and it's giving "we need a new addressing vibe" 🧚.

The wireless world is literally trying to get around the subnet model of addressing, either with proxies or with IP address servers 👀. So maybe IPng needs an addressing model that's more flexible than subnetting, both to cut down the administrative burden and to make life easier for roaming users 🌈.

The need to eliminate single points of failure drives the business requirement for multi-tail attachment of hosts to networks 🛟. Corporate users complain that TCP/IP can non-disruptively switch a connection from a broken route to a working one only if the new route leads to the same adapter on the end system 😩.

Configuration, Administration and Operation 🛠️

Businesses would like dynamic but secure updating of Domain Name Servers 🔄, both to make moves and changes easier and to facilitate cutover to backup hosts. In this vein, secure and dynamic interaction between DNS and Dynamic Host Configuration Protocol (DHCP, RFC 1541) is required 🔐. The IETF should encourage wide deployment of DHCP, and then solicit feedback on whether additional configuration requirements need to be satisfied 📝.

Policy-Based Routing 🗺️

Policy-based routing is more of a solution than a requirement 🧩. Businesses rarely require a general purpose policy architecture, although they do state requirements that policy-based routing could satisfy. For example, corporations do not want to carry for free the transit traffic of other enterprises 🙅‍♀️, and they may not want their sensitive data to flow through links controlled by certain other enterprises 🔒.

Policy-based routing is one possible way to satisfy those requirements, but there seems to be a concern that general purpose policy-based routing may have high administrative cost and low routing performance 📉.

Scaling 📈

If IPng satisfies the scaling requirement of the Internet, then it satisfies it for corporate networks a fortiori 💯.

Conclusions 👑

Enhancements to the Internet protocol suite, together with wider deployment of some of its existing architectures, could satisfy these requirements of commercial customers while retaining IPv4 🌈. Expansion of the address space eventually will be necessary to allow continued Internet growth, but in RFC 1518 Tony Li and Yakov Rehkter have shown that from a technical perspective the addressing issue of IPng is not an immediate concern 🙌.

Nevertheless, the TCP/IP community should establish a direction for enlargement of the address space, because unfounded publicity about the address space is scaring away potential TCP/IP users 😱. If the IETF does not provide direction on how its address space will grow, then people may use non-standard, and probably incompatible, approaches 🙃.

Security Considerations 🔐

The IETF should encourage wide deployment of GSS API, and then solicit feedback on whether additional security requirements need to be satisfied 🛡️. Businesses are so concerned about network cost control mechanisms that they want them secured against tampering 💰.

IPng should not interfere with firewalls, which many corporations consider essential 🔥. Don't mess with those, please 🙏.

See other comments on Security throughout this memo 🧵.

Authors' Addresses 💌

Meet the authors, bestie 🫶:

Edward Britton
IBM Corp.
E69/503
P.O.Box 12195
Research Triangle Park, NC 27709

Phone: (919) 254-6037
EMail: brittone@vnet.ibm.com


John Tavs
IBM Corp.
E69/503
P.O.Box 12195
Research Triangle Park, NC 27709

Phone: (919) 245-7610
EMail: tavs@vnet.ibm.com

📦 where this page comes from

ietf: rfc1678 1994-08
💖 girlified by claude-haiku-5-5 on 2026-10-15: a language model's rewrite of the original, under its terms