RFC 5772
A Set of Possible Requirements for a Future Routing Architecture
Pages: 68
Historic
Historic
Part 3 of 3 – Pages 42 to 68
3.6. Functional Requirements
This section includes a detailed discussion of new requirements for a Future Domain Routing architecture. The nth requirement carries the label "R(n)". As discussed in Section 3.2.3.12, a new architecture
must build upon the requirements of the past routing framework and must not reduce the functionality of the network. A discussion and analysis of the RFC 1126 requirements can be found in [RFC5773].3.6.1. Topology
3.6.1.1. Routers Should Be Able to Learn and to Exploit the Domain Topology
R(1) Routers must be able to acquire and hold sufficient information on the underlying topology of the domain to allow the establishment of routes on that topology. R(2) Routers must have the ability to control the establishment of routes on the underlying topology. R(3) Routers must be able, where appropriate, to control Sub-IP mechanisms to support the establishment of routes. The OSI Inter-Domain Routing Protocol (IDRP) [ISO10747] allowed a collection of topologically related domains to be replaced by an aggregate domain object, in a similar way to the Nimrod [Chiappa02] domain hierarchies. This allowed a route to be more compactly represented by a single collection instead of a sequence of individual domains. R(4) Routers must, where appropriate, be able to construct abstractions of the topology that represent an aggregation of the topological features of some area of the topology.3.6.1.2. The Same Topology Information Should Support Different Path Selection Ideas
The same topology information needs to provide the more flexible spectrum of path selection methods that we might expect to find in a future Internet, including distributed techniques such as hop-by-hop, shortest path, local optimization constraint-based, class of service, source address routing, and destination address routing, as well as the centralized, global optimization constraint-based "traffic engineering" type. Allowing different path selection techniques will produce a much more predictable and comprehensible result than the "clever tricks" that are currently needed to achieve the same results. Traffic engineering functions need to be combined. R(5) Routers must be capable of supporting a small number of different path selection algorithms.
3.6.1.3. Separation of the Routing Information Topology from the Data Transport Topology
R(6) The controlling network may be logically separate from the controlled network. The two functional "planes" may physically reside in the same nodes and share the same links, but this is not the only possibility, and other options may sometimes be necessary. An example is a pure circuit switch (that cannot see individual IP packets) combined with an external controller. Another example may be multiple links between two routers, where all the links are used for data forwarding but only one is used for carrying the routing session.3.6.2. Distribution
3.6.2.1. Distribution Mechanisms
R(7) Relevant changes in the state of the network, including modifications to the topology and changes in the values of dynamic capabilities, must be distributed to every entity in the network that needs them, in a reliable and trusted way, at the earliest appropriate time after the changes have occurred. R(8) Information must not be distributed outside areas where it is needed, or believed to be needed, for the operation of the routing system. R(9) Information must be distributed in such a way that it minimizes the load on the network, consistent with the required response time of the network to changes.3.6.2.2. Path Advertisement
R(10) The router must be able to acquire and store additional static and dynamic information that relates to the capabilities of the topology and its component nodes and links and that can subsequently be used by path selection methods. The inter-domain routing system must be able to advertise more kinds of information than just connectivity and domain paths. R(11) The routing system must support service specifications, e.g., the Service Level Specifications (SLSs) developed by the Differentiated Services working group [RFC3260].
Careful attention should be paid to ensuring that the distribution of
additional information with path advertisements remains scalable as
domains and the Internet get larger, more numerous, and more
diversified.
R(12) The distribution mechanism used for distributing network state
information must be scalable with respect to the expected size
of domains and the volume and rate of change of dynamic state
that can be expected.
The combination of R(9) and R(12) may result in a compromise between
the responsiveness of the network to change and the overhead of
distributing change notifications. Attempts to respond to very rapid
changes may damage the stability of the routing system.
Possible examples of additional capability information that might be
carried include:
- QoS information
To allow an ISP to sell predictable end-to-end QoS service to any
destination, the routing system should have information about the
end-to-end QoS. This means that:
R(13) The routing system must be able to support different paths for
different services.
R(14) The routing system must be able to forward traffic on the path
appropriate for the service selected for the traffic, either
according to an explicit marking in each packet (e.g., MPLS
labels, Diffserv Per-Hop Behaviors (PHBs) or DSCP values) or
implicitly (e.g., the physical or logical port on which the
traffic arrives).
R(15) The routing system should also be able to carry information
about the expected (or actually, promised) characteristics of
the entire path and the price for the service.
(If such information is exchanged at all between network operators
today, it is through bilateral management interfaces, and not
through the routing protocols.)
This would allow for the operator to optimize the choice of path
based on a price/performance trade-off.
In addition to providing dynamic QoS information, the system
should be able to use static class-of-service information.
- Security information
Security characteristics of other domains referred to by
advertisements can allow the routing entity to make routing
decisions based on political concerns. The information itself is
assumed to be secure so that it can be trusted.
- Usage and cost information
Usage and cost information can be used for billing and traffic
engineering. In order to support cost-based routing policies for
customers (i.e., peer ISPs), information such as "traffic on this
link or path costs XXX per Gigabyte" needs to be advertised, so
that the customer can choose a more or a less expensive route.
- Monitored performance
Performance information such as delay and drop frequency can be
carried. (This may only be suitable inside a domain because of
trust considerations.) This should support at least the kind of
delay-bound contractual terms that are currently being offered by
service providers. Note that these values refer to the outcome of
carrying bits on the path, whereas the QoS information refers to
the proposed behavior that results in this outcome.
- Multicast information
R(16) The routing system must provide information needed to create
multicast distribution trees. This information must be
provided for one-to-many distribution trees and should be
provided for many-to-many distribution trees.
The actual construction of distribution trees is not necessarily
done by the routing system.
3.6.2.3. Stability of Routing Information
R(17) The new network architecture must be stable without needing
global convergence, i.e., convergence is a local property.
The degree to which this is possible and the definition of "local"
remain research topics. Restricting the requirement for convergence
to localities will have an effect on all of the other requirements in
this section.
R(18) The distribution and the rate of distribution of changes must
not affect the stability of the routing information. For
example, commencing redistribution of a change before the
previous one has settled must not cause instability.
3.6.2.3.1. Avoiding Routing Oscillations
R(19) The routing system must minimize oscillations in route
advertisements.
3.6.2.3.2. Providing Loop-Free Routing and Forwarding
In line with the separation of routing and forwarding concerns:
R(20) The distribution of routing information must be, so far as is
possible, loop-free.
R(21) The forwarding information created from this routing
information must seek to minimize persistent loops in the
data-forwarding paths.
It is accepted that transient loops may occur during convergence of
the protocol and that there are trade-offs between loop avoidance and
global scalability.
3.6.2.3.3. Detection, Notification, and Repair of Failures
R(22) The routing system must provide means for detecting failures
of node equipment or communication links.
R(23) The routing system should be able to coordinate failure
indications from Layer 3 mechanisms, from nodal mechanisms
built into the routing system, and from lower-layer mechanisms
that propagate up to Layer 3 in order to determine the root
cause of the failure. This will allow the routing system to
react correctly to the failure by activating appropriate
mitigation and repair mechanisms if required, while ensuring
that it does not react if lower-layer repair mechanisms are
able to repair or mitigate the fault.
Most Layer 3 routing protocols have utilized keepalives or "hello"
protocols as a means of detecting failures at Layer 3. The keepalive
mechanisms are often complemented by analog mechanisms (e.g., laser-
light detection) and hardware mechanisms (e.g., hardware/software
watchdogs) that are built into routing nodes and communication links.
Great care must be taken to make the best possible use of the various
failure repair methods available while ensuring that only one repair
mechanism at a time is allowed to repair any given fault.
Interactions between, for example, fast reroute mechanisms at Layer 3
and Synchronous Optical Network/Synchronous Digital Hierarchy (SONET/
SDH) repair at Layer 1 are highly undesirable and are likely to cause
problems in the network.
R(24) Where a network topology and routing system contains multiple
fault repair mechanisms, the responses of these systems to a
detected failure should be coordinated so that the fault is
repaired by the most appropriate means, and no extra repairs
are initiated.
R(25) Where specialized packet exchange mechanisms (e.g., Layer 3
keepalive or "hello" protocol mechanisms) are used to detect
failures, the routing system must allow the configuration of
the rate of transmission of these keepalives. This must
include the capability to turn them off altogether for links
that are deliberately broken when no real user or control
traffic is present (e.g., ISDN links).
This will allow the operator to compromise between the speed of
failure detection and the proportion of link bandwidth dedicated to
failure detection.
3.6.3. Addressing
3.6.3.1. Support Mix of IPv4, IPv6, and Other Types of Addresses
R(26) The routing system must support a mix of different kinds of
addresses.
This mix will include at least IPv4 and IPv6 addresses, and
preferably various types of non-IP addresses, too. For instance,
networks like SDH/SONET and Wavelength Division Multiplexing (WDM)
may prefer to use non-IP addresses. It may also be necessary to
support multiple sets of "private" (e.g., RFC 1918) addresses when
dealing with multiple customer VPNs.
R(27) The routing system should support the use of a single topology
representation to generate routing and forwarding tables for
multiple address families on the same network.
This capability would minimize the protocol overhead when exchanging
routes.
3.6.3.2. Support for Domain Renumbering/Readdressing
R(28) If a domain is subject to address reassignment that would cause forwarding interruption, then the routing system should support readdressing (e.g., when a new prefix is given to an old network, and the change is known in advance) by maintaining routing during the changeover period [RFC2071] [RFC2072].3.6.3.3. Multicast and Anycast
R(29) The routing system must support multicast addressing, both within a domain and across multiple domains. R(30) The routing system should support anycast addressing within a domain. The routing system may support anycast addressing across domains. An open question is whether it is possible or useful to support anycast addressing between cooperating domains.3.6.3.4. Address Scoping
R(31) The routing system must support scoping of unicast addresses, and it should support scoping of multicast and anycast address types. The unicast address scoping that is being designed for IPv6 does not seem to cause any special problems for routing. IPv6 inter-domain routing handles only IPv6 global addresses, while intra-domain routing also needs to be aware of the scope of private addresses. Editors' Note: the original reference was to site-local addresses, but these have been deprecated by the IETF. Link-local addresses are never routed at all. More study may be needed to identify the requirements and solutions for scoping in a more general sense and for scoping of multicast and anycast addresses.3.6.3.5. Mobility Support
R(32) The routing system must support system mobility. The term "system" includes anything from an end system to an entire domain.
We observe that the existing solutions based on renumbering and/or tunneling are designed to work with the current routing, so they do not add any new requirements to future routing. But the requirement is general, and future solutions may not be restricted to the ones we have today.3.6.4. Statistics Support
R(33) Both the routing and forwarding parts of the routing system must maintain statistical information about the performance of their functions.3.6.5. Management Requirements
While the tools of management are outside the scope of routing, the mechanisms to support the routing architecture and protocols are within scope. R(34) Mechanisms to support Operational, Administrative, and Management control of the routing architecture and protocols must be designed into the original fabric of the architecture.3.6.5.1. Simple Policy Management
The basic aims of this specification are: - to require less manual configuration than today, and - to satisfy the requirements for both easy handling and maximum control. That is: - All the information should be available, - but should not be visible except for when necessary. - Policies themselves should be advertised and not only the result of policy, and - policy-conflict resolution must be provided. R(35) The routing system must provide management of the system by means of policies. For example, policies that can be expressed in terms of the business and services implemented on the network and reflect the operation of the network in terms of the services affected.
Editors' Note: This requirement is optimistic in that it
implies that it is possible to get operators to
cooperate even if it is seen by them to be against their
business practices.
R(36) The distribution of policies must be amenable to scoping to
protect proprietary policies that are not relevant beyond the
local set of domains.
3.6.5.2. Startup and Maintenance of Routers
A major problem in today's networks is the need to perform initial
configuration on routers from a local interface before a remote
management system can take over. It is not clear that this imposes
any requirements on the routing architecture beyond what is needed
for a ZeroConf host.
Similarly, maintenance and upgrade of routers can cause major
disruptions to the network routing because the routing system and
management of routers is not organized to minimize such disruption.
Some improvements have been made, such as graceful restart mechanisms
in protocols, but more needs to be done.
R(37) The routing system and routers should provide mechanisms that
minimize the disruption to the network caused by maintenance
and upgrades of software and hardware. This requirement
recognizes that some of the capabilities needed are outside
the scope of the routing architecture (e.g., minimum impact
software upgrade).
3.6.6. Provability
R(38) The routing system and its component protocols must be
demonstrated to be locally convergent under the permitted
range of parameter settings and policy options that the
operator(s) can select.
There are various methods for demonstration and proof that include,
but are not limited to: mathematical proof, heuristic, and pattern
recognition. No requirement is made on the method used for
demonstrating local convergence properties.
R(39) Routing protocols employed by the routing system and the
overall routing system should be resistant to bad routing
policy decisions made by operators.
Tools are needed to check compatibility of routing policies. While
these tools are not part of the routing architecture, the mechanisms
to support such tools are.
Routing policies are compatible if their interaction does not cause
instability. A domain or group of domains in a system is defined as
being convergent, either locally or globally, if and only if, after
an exchange of routing information, routing tables reach a stable
state that does not change until the routing policies or the topology
changes again.
To achieve the above-mentioned goals:
R(40) The routing system must provide a mechanism to publish and
communicate policies so that operational coordination and
fault isolation are possible.
Tools are required that verify the stability characteristics of the
routing system in specified parts of the Internet. The tools should
be efficient (fast) and have a broad scope of operation (check large
portions of Internet). While these tools are not part of the
architecture, developing them is in the interest of the architecture
and should be defined as a Routing Research Group activity while
research on the architecture is in progress.
Tools analyzing routing policies can be applied statically or
(preferably) dynamically. A dynamic solution requires tools that can
be used for run time checking for oscillations that arise from policy
conflicts. Research is needed to find an efficient solution to the
dynamic checking of oscillations.
3.6.7. Traffic Engineering
The ability to do traffic engineering and to get the feedback from
the network to enable traffic engineering should be included in the
future domain architecture. Though traffic engineering has many
definitions, it is, at base, another alternative or extension for the
path selection mechanisms of the routing system. No fundamental
changes to the requirements are needed, but the iterative processes
involved in traffic engineering may require some additional
capabilities and state in the network.
Traffic engineering typically involves a combination of off-line
network planning and administrative control functions in which the
expected and measured traffic flows are examined, resulting in
changes to static configurations and policies in the routing system.
During operations, these configurations control the actual flow of traffic and affect the dynamic path selection mechanisms; the results are measured and fed back into further rounds of network planning.3.6.7.1. Support for, and Provision of, Traffic Engineering Tools
At present, there is an almost total lack of effective traffic engineering tools, whether in real time for network control or off- line for network planning. The routing system should encourage the provision of such tools. R(41) The routing system must generate statistical and accounting information in such a way that traffic engineering and network planning tools can be used in both real-time and off-line planning and management.3.6.7.2. Support of Multiple Parallel Paths
R(42) The routing system must support the controlled distribution over multiple links or paths of traffic toward the same destination. This applies to domains with two or more connections to the same neighbor domain, and to domains with connections to more than one neighbor domain. The paths need not have the same metric. R(43) The routing system must support forwarding over multiple parallel paths when available. This support should extend to cases where the offered traffic is known to exceed the available capacity of a single link, and to the cases where load is to be shared over paths for cost or resiliency reasons. R(44) Where traffic is forwarded over multiple parallel paths, the routing system must, so far as is possible, avoid the reordering of packets in individual micro-flows. R(45) The routing system must have mechanisms to allow the traffic to be reallocated back onto a single path when multiple paths are not needed.3.6.7.3. Peering Support
R(46) The routing system must support peer-level connectivity as well as hierarchical connections between domains.
The network is becoming increasingly complex, with private peering
arrangements set up between providers at every level of the hierarchy
of service providers and even by certain large enterprises, in the
form of dedicated extranets.
R(47) The routing system must facilitate traffic engineering of peer
routes so that traffic can be readily constrained to travel as
the network operators desire, allowing optimal use of the
available connectivity.
3.6.8. Support for Middleboxes
One of our assumptions is that NATs and other middle-boxes such as
firewalls, web proxies, and address family translators (e.g., IPv4 to
IPv6) are here to stay.
R(48) The routing system should work in conjunction with middle-
boxes, e.g., NAT, to aid in bi-directional connectivity
without compromising the additional opacity and privacy that
the middle-boxes offer.
This problem is closely analogous to the abstraction problem, which
is already under discussion for the interchange of routing
information between domains.
3.7. Performance Requirements
Over the past several years, the performance of the routing system
has frequently been discussed. The requirements that derive from
those discussions are listed below. The specific values for these
performance requirements are left for further discussion.
R(49) The routing system must support domains of at least N systems.
A system is taken to mean either an individual router or a
domain.
R(50) Local convergence should occur within T units of time.
R(51) The routing system must be measurably reliable. The measure
of reliability remains a research question.
R(52) The routing system must be locally stable to a measured
degree. The degree of measurability remains a research issue.
R(53) The routing system must be globally stable to a measured
degree. The degree of measurability remains a research issue.
R(54) The routing system should scale to an indefinitely large
number of domains.
There has been very little data or statistical evidence for many of
the performance claims made in the past. In recent years, several
efforts have been initiated to gather data and do the analyses
required to make scientific assessments of performance issues and
requirements. In order to complete this section of the requirements
analysis, the data and analyses from these studies needs to be
gathered and collated into this document. This work has been started
but has yet to be completed.
Editors' Note: This work was never completed due to corporate
reorganizations.
3.8. Backward Compatibility (Cutover) and Maintainability
This area poses a dilemma. On one hand, it is an absolute
requirement that:
R(55) The introduction of the routing system must not require any
flag days.
R(56) The network currently in place must continue to run at least
as well as it does now while the new network is being
installed around it.
However, at the same time, it is also an requirement that:
R(57) The new architecture must not be limited by the restrictions
that plague today's network.
It has to be admitted that R(57) is not a well-defined requirement,
because we have not fully articulated what the restrictions might be.
Some of these restrictions can be derived by reading the discussions
for the positive requirements above. It would be a useful exercise
to explicitly list all the restrictions and irritations with which we
wish to do away. Further, it would be useful to determine if these
restrictions can currently be removed at a reasonable cost or whether
we are actually condemned to live with them.
Those restrictions cannot be allowed to become permanent baggage on
the new architecture. If they do, the effort to create a new system
will come to naught. It may, however, be necessary to live with some
of them temporarily for practical reasons while providing an
architecture that will eventually allow them to be removed. The last
three requirements have significance not only for the transition
strategy but also for the architecture itself. They imply that it must be possible for an internet such as today's BGP-controlled network, or one of its ASs, to exist as a domain within the new FDR.3.9. Security Requirements
As previously discussed, one of the major changes that has overtaken the Internet since its inception is the erosion of trust between end users making use of the net, between those users and the suppliers of services, and between the multiplicity of providers. Hence, security, in all its aspects, will be much more important in the FDR. It must be possible to secure the routing communication. R(58) The communicating entities must be able to identify who sent and who received the information (authentication). R(59) The communicating entities must be able to verify that the information has not been changed on the way (integrity). Security is more important in inter-domain routing where the operator has no control over the other domains, than in intra-domain routing where all the links and the nodes are under the administration of the operator and can be expected to share a trust relationship. This property of intra-domain trust, however, should not be taken for granted: R(60) Routing communications must be secured by default, but an operator must have the option to relax this requirement within a domain where analysis indicates that other means (such as physical security) provide an acceptable alternative. R(61) The routing communication mechanism must be robust against denial-of-service attacks. R(62) It should be possible to verify that the originator of the information was authorized to generate the information. Further considerations that may impose further requirements include: - whether no one else but the intended recipient is able to access (privacy) or understand (confidentiality) the information, - whether it is possible to verify that all the information has been received and that the two parties agree on what was sent (validation and non-repudiation),
- whether there is a need to separate security of routing from
security of forwarding, and
- whether traffic flow security is needed (i.e., whether there is
value in concealing who can connect to whom, and what volumes of
data are exchanged).
Securing the BGP session, as done today, only secures the exchange of
messages from the peering domain, not the content of the information.
In other words, we can confirm that the information we got is what
our neighbor really sent us, but we do not know whether or not this
information (that originated in some remote domain) is true.
A decision has to be made on whether to rely on chains of trust (we
trust our peers who trust their peers who..), or whether we also need
authentication and integrity of the information end-to-end. This
information includes both routes and addresses. There has been
interest in having digital signatures on originated routes as well as
countersignatures by address authorities to confirm that the
originator has authority to advertise the prefix. Even understanding
who can confirm the authority is non-trivial, as it might be the
provider who delegated the prefix (with a whole chain of authority
back to ICANN) or it may be an address registry. Where a prefix
delegated by a provider is being advertised through another provider
as in multi-homing, both may have to be involved to confirm that the
prefix may be advertised through the provider who doesn't have any
interest in the prefix!
R(63) The routing system must cooperate with the security policies
of middle-boxes whenever possible.
This is likely to involve further requirements for abstraction of
information. For example, a firewall that is seeking to minimize
interchange of information that could lead to a security breach. The
effect of such changes on the end-to-end principle should be
carefully considered as discussed in [Blumenthal01].
R(64) The routing system must be capable of complying with local
legal requirements for interception of communication.
3.10. Debatable Issues
This section covers issues that need to be considered and resolved in
deciding on a Future Domain Routing architecture. While they can't
be described as requirements, they do affect the types of solution
that are acceptable. The discussions included below are very open-
ended.
3.10.1. Network Modeling
The mathematical model that underlies today's routing system uses a graph representation of the network. Hosts, routers, and other processing boxes are represented by nodes and communications links by arcs. This is a topological model in that routing does not need to directly model the physical length of the links or the position of the nodes; the model can be transformed to provide a convenient picture of the network by adjusting the lengths of the arcs and the layout of the nodes. The connectivity is preserved and routing is unaffected by this transformation. The routing algorithms in traditional routing protocols utilize a small number of results from graph theory. It is only recently that additional results have been employed to support constraint-based routing for traffic engineering. The naturalness of this network model and the "fit" of the graph theoretical methods may have tended to blind us to alternative representations and inhibited us from seeking alternative strands of theoretical thinking that might provide improved results. We should not allow this habitual behavior to stop us from looking for alternative representations and algorithms; topological revolutions are possible and allowed, at least in theory.3.10.2. System Modeling
The assumption that object modeling of a system is an essential first step to creating a new system is still novel in this context. Frequently, the object modeling effort becomes an end in itself and does not lead to system creation. But there is a balance, and a lot that can be discovered in an ongoing effort to model a system such as the Future Domain Routing system. It is recommended that this process be included in the requirements. It should not, however, be a gating event to all other work. Some of the most important realizations will occur during the process of determining the following: - Object classification - Relationships and containment - Roles and Rules
3.10.3. One, Two, or Many Protocols
There has been a lot of discussion of whether the FDR protocol solution should consist of one (probably new) protocol, two (intra- and inter-domain) protocols, or many protocols. While it might be best to have one protocol that handles all situations, this seems improbable. On the other hand, maintaining the "strict" division evident in the network today between the IGP and EGP may be too restrictive an approach. Given this, and the fact that there are already many routing protocols in use, the only possible answer seems to be that the architecture should support many protocols. It remains an open issue, one for the solution, to determine if a new protocol needs to be designed in order to support the highest goals of this architecture. The expectation is that a new protocol will be needed.3.10.4. Class of Protocol
If a new protocol is required to support the FDR architecture, the question remains open as to what kind of protocol this ought to be. It is our expectation that a map distribution protocol will be required to augment the current path-vector protocol and shortest path first protocols.3.10.5. Map Abstraction
Assuming that a map distribution protocol, as defined in [RFC1992] is required, what are the requirements on this protocol? If every detail is advertised throughout the Internet, there will be a lot of information. Scalable solutions require abstraction. - If we summarize too much, some information will be lost on the way. - If we summarize too little, then more information than required is available, contributing to scaling limitations. - One can allow more summarization, if there also is a mechanism to query for more details within policy limits. - The basic requirement is not that the information shall be advertised, but rather that the information shall be available to those who need it. We should not presuppose a solution where advertising is the only possible mechanism.
3.10.6. Clear Identification for All Entities
As in all other fields, the words used to refer to concepts and to describe operations about routing are important. Rather than describe concepts using terms that are inaccurate or rarely used in the real world of networking, it is necessary to make an effort to use the correct words. Many networking terms are used casually, and the result is a partial or incorrect understanding of the underlying concept. Entities such as nodes, interfaces, subnetworks, tunnels, and the grouping concepts such as ASs, domains, areas, and regions, need to be clearly identified and defined to avoid confusion. There is also a need to separate identifiers (what or who) from locators (where) from routes (how to reach). Editors' Note: Work was undertaken in the shim6 working group of the IETF on this sort of separation. This work needs to be taken into account in any new routing architecture.3.10.7. Robustness and Redundancy
The routing association between two domains should survive even if some individual connection between two routers goes down. The "session" should operate between logical "routing entities" on each domain side, and not necessarily be bound to individual routers or addresses. Such a logical entity can be physically distributed over multiple network elements. Or, it can reside in a single router, which would default to the current situation.3.10.8. Hierarchy
A more flexible hierarchy with more levels and recursive groupings in both upward and downward directions allows more structured routing. The consequence is that no single level will get too big for routers to handle. On the other hand, it appears that the real-world Internet is becoming less hierarchical, so that it will be increasingly difficult to use hierarchy to control scaling. Note that groupings can look different depending on which aspect we use to define them. A Diffserv area, an MPLS domain, a trusted domain, a QoS area, a multicast domain, etc., do not always coincide; nor are they strict hierarchical subsets of each other. The basic distinction at each level is "this grouping versus everything outside".
3.10.9. Control Theory
Is it possible to apply a control theory framework to analyze the stability of the control system of the whole network domain, with regard to, e.g., convergence speed and the frequency response, and then use the results from that analysis to set the timers and other protocol parameters? Control theory could also play a part in QoS routing, by modifying current link-state protocols with link costs dependent on load and feedback. Control theory is often used to increase the stability of dynamic systems. It might be possible to construct a new, totally dynamic routing protocol solely on a control theoretic basis, as opposed to the current protocols that are based in graph theory and static in nature.3.10.10. Byzantium
Is solving the Byzantine Generals problem a requirement? This is the problem of reaching a consensus among distributed units if some of them give misleading answers. The current intra-domain routing system is, at one level, totally intolerant of misleading information. However, the effect of different sorts of misleading or incorrect information has vastly varying results, from total collapse to purely local disconnection of a single domain. This sort of behavior is not very desirable. There are, possibly, other network robustness issues that must be researched and resolved.3.10.11. VPN Support
Today, BGP is also used for VPNs, for example, as described in RFC 4364 [RFC4364]. Internet routing and VPN routing have different purposes and most often exchange different information between different devices. Most Internet routers do not need to know VPN-specific information. The concepts should be clearly separated. But when it comes to the mechanisms, VPN routing can share the same protocol as ordinary Internet routing; it can use a separate instance of the same protocol or it can use a different protocol. All variants are possible and have their own merits. These requirements are silent on this issue.
3.10.12. End-to-End Reliability
The existing Internet architecture neither requires nor provides end- to-end reliability of control information dissemination. There is, however, already a requirement for end-to-end reliability of control information distribution, i.e., the ends of the VPN established need to have an acknowledgment of the success in setting up the VPN. While it is not necessarily the function of a routing architecture to provide end-to-end reliability for this kind of purpose, we must be clear that end-to-end reliability becomes a requirement if the network has to support such reliable control signaling. There may be other requirements that derive from requiring the FDR to support reliable control signaling.3.10.13. End-to-End Transparency
The introduction of private addressing schemes, Network Address Translators, and firewalls has significantly reduced the end-to-end transparency of the network. In many cases, the network is also no longer symmetric, so that communication between two addresses is possible if the communication session originates from one end but not from the other. This impedes the deployment of new peer-to-peer services and some "push" services where the server in a client- server arrangement originates the communication session. Whether a new routing system either can or should seek to restore this transparency is an open issue. A related issue is the extent to which end-user applications should seek to control the routing of communications to the rest of the network.4. Security Considerations
We address security issues in the individual requirements. We do require that the architecture and protocols developed against this set of requirements be "secure". Discussion of specific security issues can be found in the following sections: o Group A: Routing System Security - Section 2.1.9 o Group A: End Host Security - Section 2.1.10 o Group A: Routing Information Policies - Section 2.1.11.1 o Group A: Abstraction - Section 2.1.16 o Group A: Robustness - Section 2.1.18
o Group B: Protection against Denial-of-Service and Other Security
Attacks - Section 3.2.3.8
o Group B: Commercial Service Providers - Section 3.3.1.1
o Group B: The Federated Environment - Section 3.4.1
o Group B: Path Advertisement - Section 3.6.2.2
o Group B: Security Requirements - Section 3.9
5. IANA Considerations
This document is a set of requirements from which a new routing and
addressing architecture may be developed. From that architecture, a
new protocol, or set of protocols, may be developed.
While this note poses no new tasks for IANA, the architecture and
protocols developed from this document probably will have issues to
be dealt with by IANA.
6. Acknowledgments
This document is the combined effort of two groups in the IRTF.
Group A, which was formed by the IRTF Routing Research chairs, and
Group B, which was self-formed and later was folded into the IRTF
Routing Research Group. Each group has it own set of
acknowledgments.
Group A Acknowledgments
This originated in the IRTF Routing Research Group's sub-group on
Inter-domain routing requirements. The members of the group were:
Abha Ahuja Danny McPherson
J. Noel Chiappa David Meyer
Sean Doran Mike O'Dell
JJ Garcia-Luna-Aceves Andrew Partan
Susan Hares Radia Perlman
Geoff Huston Yakov Rehkter
Frank Kastenholz John Scudder
Dave Katz Curtis Villamizar
Tony Li Dave Ward
We also appreciate the comments and review received from Ran
Atkinson, Howard Berkowitz, Randy Bush, Avri Doria, Jeffery Haas,
Dmitri Krioukov, Russ White, and Alex Zinin. Special thanks to
Yakov Rehkter for contributing text and to Noel Chiappa.
Group B Acknowledgments
The document is derived from work originally produced by Babylon.
Babylon was a loose association of individuals from academia,
service providers, and vendors whose goal was to discuss issues in
Internet routing with the intention of finding solutions for those
problems.
The individual members who contributed materially to this document
are: Anders Bergsten, Howard Berkowitz, Malin Carlzon, Lenka Carr
Motyckova, Elwyn Davies, Avri Doria, Pierre Fransson, Yong Jiang,
Dmitri Krioukov, Tove Madsen, Olle Pers, and Olov Schelen.
Thanks also go to the members of Babylon and others who did
substantial reviews of this material. Specifically, we would like
to acknowledge the helpful comments and suggestions of the
following individuals: Loa Andersson, Tomas Ahlstrom, Erik Aman,
Thomas Eriksson, Niklas Borg, Nigel Bragg, Thomas Chmara, Krister
Edlund, Owe Grafford, Torbjorn Lundberg, Jeremy Mineweaser,
Jasminko Mulahusic, Florian-Daniel Otel, Bernhard Stockman, Tom
Worster, and Roberto Zamparo.
In addition, the authors are indebted to the folks who wrote all
the references we have consulted in putting this paper together.
This includes not only the references explicitly listed below, but
also those who contributed to the mailing lists we have been
participating in for years.
The editors thank Lixia Zhang, as IRSG document shepherd, for her
help and her perseverance, without which this document would never
have been published.
Finally, it is the editors who are responsible for any lack of
clarity, any errors, glaring omissions or misunderstandings.
7. Informative References
[Blumenthal01] Blumenthal, M. and D. Clark, "Rethinking the design of the Internet: The end to end arguments vs. the brave new world", May 2001, <http://dspace.mit.edu/handle/1721.1/1519>. [Broido02] Broido, A., Nemeth, E., Claffy, K., and C. Elves, "Internet Expansion, Refinement and Churn", February 2002. [CIDR] Telcordia Technologies, "CIDR Report", <http://www.cidr-report.org/>. [Chiappa02] Chiappa, N., "A New IP Routing and Addressing Architecture", July 1991, <http://ana-3.lcs.mit.edu/~jnc/nimrod/overview.txt>. [Clark91] Clark, D., "Quote reportedly from IETF Plenary discussion", 1991. [DiffservAR] Seddigh, N., Nandy, B., and J. Heinanen, "An Assured Rate Per-Domain Behaviour for Differentiated Services", Work in Progress, July 2001. [DiffservVW] Jacobson, V., Nichols, K., and K. Poduri, "The 'Virtual Wire' Per-Domain Behavior", Work in Progress, July 2000. [Griffin99] Griffin, T. and G. Wilfong, "An Analysis of BGP Convergence Properties", SIGCOMM 1999. [ISO10747] ISO/IEC, "Protocol for Exchange of Inter-Domain Routeing Information among Intermediate Systems to Support Forwarding of ISO 8473 PDUs", International Standard 10747 ISO/IEC JTC 1, Switzerland, 1993. [InferenceSRLG] Papadimitriou, D., Poppe, F., J. Jones, J., S. Venkatachalam, S., S. Dharanikota, S., Jain, R., Hartani, R., and D. Griffith, "Inference of Shared Risk Link Groups", Work in Progress, November 2001.
[ODell01] O'Dell, M., "Private Communication", 2001. [RFC1126] Little, M., "Goals and functional requirements for inter- autonomous system routing", RFC 1126, October 1989. [RFC1726] Partridge, C. and F. Kastenholz, "Technical Criteria for Choosing IP The Next Generation (IPng)", RFC 1726, Dec 1994. [RFC1992] Castineyra, I., Chiappa, N., and M. Steenstrup, "The Nimrod Routing Architecture", RFC 1992, August 1996. [RFC2071] Ferguson, P. and H. Berkowitz, "Network Renumbering Overview: Why would I want it and what is it anyway?", RFC 2071, January 1997. [RFC2072] Berkowitz, H., "Router Renumbering Guide", RFC 2072, January 1997. [RFC3031] Rosen, E., Viswanathan, A., and R. Callon, "Multiprotocol Label Switching Architecture", RFC 3031, January 2001. [RFC3221] Huston, G., "Commentary on Inter-Domain Routing in the Internet", RFC 3221, December 2001. [RFC3260] Grossman, D., "New Terminology and Clarifications for Diffserv", RFC 3260, April 2002. [RFC3344] Perkins, C., "IP Mobility Support.", RFC 3344, August 2002. [RFC3345] McPherson, D., Gill, V., Walton, D., and A. Retana, "Border Gateway Protocol (BGP) Persistent Route Oscillation Condition", RFC 3345, August 2002. [RFC3471] Berger, L., "Generalized Multi-Protocol Label Switching (GMPLS) Signaling Functional Description", RFC 3471, January 2003. [RFC3963] Devarapalli, V., Wakikawa, R., Petrescu, A., and P. Thubert, "Network Mobility (NEMO) Basic Support Protocol", RFC 3963, January 2005. [RFC4364] Rosen, E. and Y. Rekhter, "BGP/MPLS IP Virtual Private Networks (VPNs)", RFC 4364, February 2006. [RFC5773] Davies, E. and A. Doria, "Analysis of Inter-Domain Routing Requirements and History", RFC 5773, February 2010.
[Wroclawski95] Wroclowski, J., "The Metanet White Paper - Workshop on Research Directions for the Next Generation Internet", 1995. [netconf-charter] Internet Engineering Task Force, "IETF Network Configuration working group", 2005, <http://www.ietf.org/html.charters/netconf-charter.html>. [policy-charter02] Internet Engineering Task Force, "IETF Policy working group", 2002, <http://www.ietf.org/html.charters/OLD/ policy-charter.html>. [rap-charter02] Internet Engineering Task Force, "IETF Resource Allocation Protocol working group", 2002, <http://www.ietf.org/html.charters/OLD/rap-charter.html>. [snmpconf-charter02] Internet Engineering Task Force, "IETF Configuration management with SNMP working group", 2002, <http:// www.ietf.org/html.charters/OLD/snmpconf-charter.html>.
Authors' Addresses
Avri Doria LTU Lulea 971 87 Sweden Phone: +46 73 277 1788 EMail: avri@ltu.se Elwyn B. Davies Folly Consulting Soham, Cambs UK Phone: +44 7889 488 335 EMail: elwynd@dial.pipex.com Frank Kastenholz BBN Technologies 10 Moulton St. Cambridge, MA 02183 USA Phone: +1 617 873 8047 EMail: frank@bbn.com