Network Working Group B. Zhang
Internet-Draft Univ. of Arizona
Intended status: Informational L. Zhang
Expires: April 29, 2010 UCLA
L. Wang
Univ. of Memphis
October 26, 2009
Evolution Towards Global Routing Scalability
draft-zhang-zhang-evolution-02.txt
Status of this Memo
This Internet-Draft is submitted to IETF in full conformance with the
provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet-
Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
This Internet-Draft will expire on April 29, 2010.
Copyright Notice
Copyright (c) 2009 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents in effect on the date of
publication of this document (http://trustee.ietf.org/license-info).
Please review these documents carefully, as they describe your rights
and restrictions with respect to this document.
This document may contain material from IETF Documents or IETF
Contributions published or made publicly available before November
Zhang, et al. Expires April 29, 2010 [Page 1]
Internet-Draft Scaling BGP October 2009
10, 2008. The person(s) controlling the copyright in some of this
material may not have granted the IETF Trust the right to allow
modifications of such material outside the IETF Standards Process.
Without obtaining an adequate license from the person(s) controlling
the copyright in such materials, this document may not be modified
outside the IETF Standards Process, and derivative works of it may
not be created outside the IETF Standards Process, except to format
it for publication as an RFC or to translate it into languages other
than English.
Abstract
Internet routing scalability has long been considered a serious
problem. Although many efforts have been devoted to address this
problem over the years, the IETF community as a whole is yet to
achieve a shared understanding on what is the best way forward. In
this draft, we step up a level to re-examine the problem and the
ongoing efforts. we conclude that, to effectively solve the routing
scalability problem, we first need a clear understanding on how to
introduce solutions to the Internet which is a global scale deployed
system. In this draft we sketch out our reasoning on the need for an
evolutionary path towards scaling the global routing system, instead
of attempting to introduce a brand new design.
Zhang, et al. Expires April 29, 2010 [Page 2]
Internet-Draft Scaling BGP October 2009
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4
2. Difficulties in Deploying New Solutions . . . . . . . . . . . 5
3. An Evolutionary Path towards Scalable Routing . . . . . . . . 7
3.1. Step One: Local FIB Size Reduction . . . . . . . . . . . . 7
3.2. Step Two: Network-Coordinated FIB Size Reduction . . . . . 8
3.3. Step Three: Reducing Adjacent AS Virtual Aggregation
Overhead . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.4. Step Four: Reducing RIB Size . . . . . . . . . . . . . . . 11
3.5. Step Five: Insulating the Core from Edge Churns . . . . . 12
3.6. Summary . . . . . . . . . . . . . . . . . . . . . . . . . 13
4. Evolution versus Incremental Deployability . . . . . . . . . . 14
5. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 15
6. Security Considerations . . . . . . . . . . . . . . . . . . . 15
7. Informative References . . . . . . . . . . . . . . . . . . . . 16
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 17
Zhang, et al. Expires April 29, 2010 [Page 3]
Internet-Draft Scaling BGP October 2009
1. Introduction
Internet routing scalability has long been an outstanding problem.
Over the years many efforts, including our own, have been devoted to
solve this problem. Since the 2006 IAB Workshop on Internet Routing
and Addressing [RFC4984], new IRTF/IETF efforts have been devoted to
developing a scalable routing architecture, and a number of proposals
have been put on the table [RRG]. We contributed a new design dubbed
APT [APT]; another new design LISP already has running code [LISP].
Yet no clear consensus has emerged in the community as to what is the
best way forward.
Assuming the routing scalability problem is real and we can find a
new design that is technically sound, why is it so difficult to agree
on deploying a new design that can solve the problem? We put in the
effort to understand fundamental roadblocks in rolling out APT[APT],
and came to a new understanding of the problem at hand: when facing a
problem, as engineers we naturally tend to design a new system to
solve the problem, hoping that the new design would be rolled out to
replace the old problematic one. This kind of "solution by new
design" approach can be effective in solving problems in small scale
(e.g. one could easily replace an old computer with a new one), but
it does not work for the deployed Internet. Instead, the Internet-
scale systems need to resolve problems through an evolutionary path,
not a revolutionary new design.
In this draft we first discuss the major difficulties in rolling out
a new design to solve the global routing scalability problem. These
difficulties suggest that the Internet routing infrastructure needs
an evolutionary path to move forward. We then show evidences that
such an evolutionary path indeed exists. To address the concern
about getting stuck at "local optimal" with incremental changes, we
sketch out a solution scenario which demonstrates the feasibility of
moving the routing system towards a scalable architecture through
incremental steps (see Section 3). This evolutionary path is driven
by the most severe aspect of the routing scalability problem that
each individual ISP faces at each stage, yet an interesting outcome
is that the overall routing architecture evolves towards the
separation between customer networks and provider networks, on which
our APT design was based. The major difference between the
evolutionary design and the initial APT design is that the separation
is not a starting point of the design, but rather a natural result of
the evolutionary design.
We also draw a distinction between an evolutionary process towards a
final solution direction versus the "incremental deployability" that
many previously proposed new designs claim to have (see Section 4).
One must not mistake co-existence between a new design and the
Zhang, et al. Expires April 29, 2010 [Page 4]
Internet-Draft Scaling BGP October 2009
existing system as incremental deployability, because the latter
requires offering immediate gains for first movers, otherwise no one
has incentive to be first movers.
2. Difficulties in Deploying New Solutions
Two of the few fundamental properties of the Internet are its
distributed governance and its diversity along multiple dimensions.
The Internet is an interconnect of tens of thousands independently
administrated networks, each with its own budget, planning, business
models and operational practices. As a result, not everyone shares
the same view as far as the routing scalability issue is concerned.
For example, many customer networks and small regional providers do
not carry the full BGP routing table internally; instead they
propagate only internal routes inside their networks and use default
routing to reach the rest of the Internet through one or a few exit
points. On the other hand, large networks in general carry the full
routing table internally for efficient data delivery to a large
number of destinations. As a result, the former may not feel the
pain of routing table growth but the latter may do.
Even among the networks that do carry the full routing table
internally, some (such as content providers) are able to upgrade
their routing infrastructure every few years to keep up with the
demand of ever growing BGP table; others may not be able to afford
doing so. For example, we learned from a few large ISPs that,
although they were able to upgrade the relatively small number of
core routers with the latest technology that can handle a million or
more routes, they could not afford to upgrade all their edge routers
which may count up to a thousand or more, even though some of them
are more than 10 years old. Consequently, some networks may
encounter the FIB or RIB size limitations earlier than others, some
may experience severe problems while others may not feel the problem
at all. Even within the same network, some routers can handle the
increasing routing table size while others cannot. Several incidents
have occurred recently that were caused by edge router RIB or FIB
overflow. Although these incidents may be triggered by other
problems (e.g., route leak-out) that led to the inflation of the RIB
size, they did show the fact that a large RIB size can easily push
old edge routers to fall off the cliff.
Therefore, although finding a way to control routing table size is
necessary in the long run, especially in lieu of increasing IPv6
deployment, different networks can have different degrees of
incentive to solve the problem, and some may not see a need to take
any action towards fixing the problem for the time being.
Zhang, et al. Expires April 29, 2010 [Page 5]
Internet-Draft Scaling BGP October 2009
Yet another important issue in solution evaluation is network
economics. A new solution design usually calls for software upgrade
or even new hardware, both require additional investment as well as
new expertise in managing and troubleshooting the new technology.
The affordability associated with deploying a new design varies
greatly among different networks. Even if a network may suffer pain
from the growing routing table size, it still may not be able to
deploy a new solution if the cost is considered prohibitively high.
Instead, people tend to look for simple twists of the existing
systems that can provide effective relief from the RIB/FIB growth
pressure. One such simple patch was presented at October 2008 NANOG
meeting [NANOG44].
Each network makes its own business decision on whether to deploy a
new design, based on its evaluation of the severity of the problem
and its affordability of deploying the solution. Given the scale and
diversity of the Internet, it is certain that the buy-in of any new
solution will not be harmonious. Even for those networks that
require a solution to handle routing scalability, the deployment will
likely be a gradual process consisting of several stages.
Furthermore, the day for the global Internet as a whole to deploy a
new solution may take forever to come.
To summarize: we see that
o Different parties have different perceptions regarding the routing
scalability problem due to their differences in economical
conditions and operational practices; some are yet to be convinced
that the routing scalability problem is serious [BGP2008].
o For networks that face the routing scalability problem, there can
still be different severity at different routers.
o Networks that experience routing scalability problems are also
likely to have affordability concerns for new solutions.
o If any new solution gets rolled out, it is certain to start from
one or a few parties first, and may or may not ever reach the
entire Internet.
The above argues that we should attack the routing scalability
problem with an evolutionary approach. By evolution we mean that (1)
the solution should be deployable by individual AS who deems an
action necessary, without needing coordination with neighbor ASes;
(2) the solution can bring immediate gain to a single first-mover;
(3) even within the AS, the solution should enable the routing table
size reduction at only those routers whose capacity fall behind the
the FIB or RIB growth curve; and (4) the solution should be an
incremental step on top of the existing system to minimize the cost
while being effective. Building a solution on top of the existing
system makes it much cheaper and easier to roll out, and makes it
Zhang, et al. Expires April 29, 2010 [Page 6]
Internet-Draft Scaling BGP October 2009
likely to work transparently with the rest of the system that does
not make the changes or does not make the change at the same time.
3. An Evolutionary Path towards Scalable Routing
Based on our current understanding of the problem and the solution
space, in this section we sketch out an evolutionary path towards
scaling Inter-domain routing. As the Internet continues to evolve
over time, it is likely that our understanding will also evolve, thus
the specific path we sketched out in this draft may, or is likely to,
change. The main point we argue in this draft is not any specific
evolutionary path that the Internet may take, but rather we aim to
show strong evidences that such an evolutionary path both exists and
is feasible; that we should aim for an evolutionary path to address
routing scalability problem, rather than attempting a brand new
design; and, most of all, that such incremental steps should not
result in "local maximal" situation, instead they can indeed move the
overall system towards the routing architecture that our new design,
APT, aimed for.
At this time we can see several steps in evolving today's BGP routing
system towards a controllable growth of the routing table size. We
identify potentially most severe pain at each step that warrants a
fix. We then identify a fix that has a reasonable cost, can be
carried out by individual networks, and can be built on top of the
existing operations, so that it does not break any other parts of the
global routing system. Note that any such simple fix necessarily has
its limitations. As the fix gets widely deployed, its limitations
are likely to become more pronounced, and can become the next problem
to address. At the same time, other aspects of the routing
scalability problems that were not addressed by these fixes may
become more severe. These issues will lead to the next step of
evolving the system forward.
3.1. Step One: Local FIB Size Reduction
During early 2009 we conducted a quick survey on routing scalability
among a small group of people with operational expertise. The
results identified the fast growing FIB size as the highest priority
concern in routing scalability; this is also consistent with the
results from the IAB 2006 workshop on Routing and Addressing
[RFC4984]. Therefore, we consider reducing FIB size the first issue
to addres, and we believe there is no major disagreement regarding
this problem statement.
The proposed solutions for resolving this FIB scalability problem, on
the other hand, differ significantly. Most of the proposals
Zhang, et al. Expires April 29, 2010 [Page 7]
Internet-Draft Scaling BGP October 2009
presented to the IRTF Routing Research Group (including our own
earlier work, APT) took on the direction of a basic architectural
change. Not only is an architectural change likely to take long to
go through the IETF standardization process as well as costly to roll
out, but also it suffers from a more fundamental problem: the
difficulty to bring immediate benefits to first movers and to be
compatible with today's deployed base. We will discuss more about
the difficulties in deploying a new design with architectural changes
in Section 4.
A different direction to reduce FIB has also be proposed. Proposals
in this direction do not propose immediate architectual changes,
instead they take pragmatic approaches to reducing FIB size. A
simple idea to compress FIB size has been suggested by multiple
people independently for some time. It works in the following way:
if all longer prefixes, say those under 1.0.0.0/8, share the same
next hop with their covering prefix 1.0.0.0/8, then only 1.0.0.0/8
needs to be installed in the FIB. A recent study [FIBAggregate]
refined the basic idea to an effective FIB Aggregation scheme, and
proposed an efficient FIB update mechanism when the next hop of
either the covering or some covered prefix(es) changes and when
prefixes need to be added or removed from the FIB. Preliminary
evaluation shows that different FIB Aggregation techniques can reduce
the FIB size by 50% to 70% or more with no impact on the correctness
of packet forwarding.
FIB aggregation requires no protocol changes. However the
effectiveness of FIB aggregation depends on the aggregatability of
covered and covering prefixes, hence has a lower bound on how much it
can reduce FIB size (i.e. it cannot reduce FIB to an arbitrarily
small size). Another proposal for FIB size reduction is Virtual
Aggregation (VA) by Francis and Xu [Virtual_Aggregation]. VA has the
potential to reduce the FIB size by a factor of 10 or more.
3.2. Step Two: Network-Coordinated FIB Size Reduction
Briefly, Virtual Aggregation works as follows. An ISP can reduce its
routers FIB size by configuring a router, dubbed Aggregation Point
Router (APR), to announce a short prefix, say 1.0.0.0/8, into its own
network in place of multiple longer prefixes that fall within
1.0.0.0/8. This short prefix is called a virtual prefix. The APR
maintains FIB entries for all the longer prefixes (e.g., 1.1.0.0/16)
covered by the virtual prefix it announces, while other routers in
the network only maintain the virtual prefix 1.0.0.0/8. When a
router R receives a packet to be forwarded to 1.1.0.0/16, R's FIB
will direct the packet to the APR, and the APR then encapsulates the
packet to the egress router for the actual prefix 1.1.0.0./16.
Zhang, et al. Expires April 29, 2010 [Page 8]
Internet-Draft Scaling BGP October 2009
Both FIB Aggregation and Virtual Aggregation represent evolutionary
steps towards scaling the FIB size. Each of them can be done by an
individual ISP to effectively shrink the FIB size for its routers,
and makes no impact on the routing operations of any other networks.
Virtual Aggregation can be more effective in FIB size reduction,
however because all packets destined to the prefixes that have been
aggregated will go through the APR, Virtual Aggregation introduces
additional delivery delay (i.e., path stretch), encapsulation
overhead, as well as the potential of the APR becoming a traffic load
concentration point. Several operational steps can be applied to
mitigate these problems.
o Do not aggregate prefixes that carry heavy volumes of traffic
(popular prefixes). Based on the assumption that most traffic
falls into a small percentage of prefixes, avoiding aggregation of
popular prefixes can prevent majority of data traffic from path
stretch and prevent the APR from overload.
o One can also control APR load by using more APRs to share the
load.
o Proper positioning of APRs can minimize the path stretch
[VA_performance].
o Finally, if an APR receives heaving volume of traffic from certain
ingress router, the APR can send to this ingress router the FIB
entries that its traffic are destined to, so that the ingress
router can cache the FIB entries and encapsulate the packets
towards the egress routers directly. This will both reduce the
APR load and eliminate the path stretch.
This last technique makes an APR perform more or less in the same way
as a Default Mapper (DM) in our APT design [APT], however with one
fundamental difference. Deploying an APR does not necessarily
require any new protocol or a new functional box (the DM node) that
the APT deployment would require. Instead, an operator can simply
configure a router to be an APR. Only when the APR rollout becomes
successful and the APR load becomes an issue, then the operator may
consider additional changes to make the ingress routers handle
caching.
We believe that FIB Aggregation (FA) and Virtual Aggregation (VA) can
be deployed sequentially or in parallel to effectively reduce the FIB
size at most routers. Virtual Aggregation can be viewed as a poor
man's map-encap within one AS. The APR holds the mapping table from
the virtual prefix to all the egress routers through which the
specific prefixes can be reached. This mapping information is
directly derived from BGP routing updates without a new mapping
distribution system. The APR then encapsulates packets to those
egress routers.
Zhang, et al. Expires April 29, 2010 [Page 9]
Internet-Draft Scaling BGP October 2009
How much time can FA and VA buy us in curtailing the FIB size growth?
It seems only time can tell. But if we look ahead one step, as the
Internet continues to grow, and as IPv6 deployment starts rolling
out, more networks may face the FIB size problem and adopt FA and VA
as solutions. When two or more adjacent ASes all deploy Virtual
Aggregation, packets that traverse these ASes will experience the
cumulated path stretches and encapsulation/decapsulation cost of all
the ASes along their paths. The need to resolve this new problem (of
cumulated path stretch and overhead) can naturally lead to the next
step of evolution towards better routing scalability.
3.3. Step Three: Reducing Adjacent AS Virtual Aggregation Overhead
Assuming the AS path a packet takes is W-X-Y-Z, and both X and Y have
deployed Virtual Aggregation. Then instead of X's own egress router,
we would like to see that X's APR encapsulates the packet directly to
the egress router of Y that connects Y to Z. This will reduce the
path stretch and the packet will only need to be encapsulated/
decapsulated once instead of two times.
To enable such inter-AS Virtual Aggregation, X's APR needs to know
Y's egress router for a given destination prefix P. This mapping
information (i.e., mapping from a destination prefix P to an egress
router) needs to be propagated from Y to X. The least resistant
approach is to piggyback such mapping information on the existing BGP
announcement for prefix P. Francis and Xu have proposed such an
extension to BGP, which carries the mapping information in a new BGP
attribute [InterDomainVA]; the APT team was also looking into more or
less the same design when the above mentioned draft was published.
We show the feasibility of this second step by the following
reasoning. First, this second step towards better routing
scalability will take place only after at least two adjacent networks
(X and Y in our example) have deployed VA and benefited from it.
Therefore we reason that they would not want to move away from VA but
would like to minimize VA's cost in path stretch and encapsulation,
to improve the performance for their customers. Second, the required
BGP implementation changes are backward compatible, meaning that
networks that have deployed this solution can easily interwork with
networks that have not deployed this solution. Furthermore, adjacent
VA-enabled ASes may not need to exchange the mapping for all of their
prefixes. Again if we take the common assumption that majority of
traffic falls within a relatively small number of prefixes, then AS Y
may only need to send AS X the mapping for a small number of prefixes
to improve the performance for a bulk part of the traffic.
As a side note we would also like to point out that this virtual
aggregation mapping exchange *closely* resembles the early design of
Zhang, et al. Expires April 29, 2010 [Page 10]
Internet-Draft Scaling BGP October 2009
APT mapping information exchange between Default Mappers [APT-00].
The content of the mapping exchanges is somewhat different, but a
more fundamental difference between what we discussed in this section
and that early APT design back in 2007 is the following: here we
sketch out an evolutionary path forward, which does not require, as a
starting point, any protocol change or information exchange across
multiple ASes that the early APT design does. Rather, the need for
mapping exchange arises only after the FIB size reduction has been
achieved, and the mapping exchange can start with two adjacent ASes
after each of them has deployed Virtual Aggregation.
3.4. Step Four: Reducing RIB Size
Piggybacking the virtual aggregation mapping information on BGP can
work well when the mapping table is small. When more networks have
adopted Virtual Aggregation, the mapping table is likely to grow
large, which may make it no longer feasible to piggyback all the
mapping information on the existing BGP sessions. The main problem,
as we can perceive today, would be the RIB size growth: A BGP router
will receive the same mapping information from multiple neighboring
BGP routers, and store all of it in its Adj-RIBs-IN. Thus BGP
routers may end up with storing multiple copies of the same mapping
information. For example assuming ISP ASes W, X, Y, and Z have a
full-mesh connectivity among themselves, and AS-W propagates a
mapping entry [eggress router R, customer prefix P], then X will
receive 3 copies of this mappy entry from Y, Z, and W, respectively.
This issue was pointed out back in 2007 when the early APT design was
discussed, and one suggestion to get around the problem is to use
separate BGP sessions for mapping information exchange. Since the
mapping information is global in scope (i.e. the pair of [eggress
router R, customer prefix P] is independent from which path one uses
to reach egress router R), this separate session can apply certain
special rules to remove duplicate entries.
Another factor is that, after a network X has deployed virtual
aggregation for a while and has gained sufficient operational
experience, it may become clear that many of its routers no longer
need to keep the full RIB table. If an internal router has small FIB
and relies on APRs to route packets towards all other destinations,
it does not need a full RIB to build its FIB. Theoretically
speaking, all border routers of X that connect to legacy networks
(i.e., those that have not deployed VA) would still need to keep the
full RIB in order to make BGP announcements into the legacy
neighbors. However in practice, only the customer-facing border
routers need a full RIB. The other border routers, those that face
either peer or provider legacy neighbors, only need to announce X's
own customer prefixes to them. Careful engineering analysis and
Zhang, et al. Expires April 29, 2010 [Page 11]
Internet-Draft Scaling BGP October 2009
configuration can eliminate the need for many routers to keep full
RIB; among those that keep the full RIB will be the ones serving as
APRs.
As we perceive what may happen further into the future, the picture
becomes more blurry, hence what we try to forecast here may or may
not bear great accuracy for what may happen in the future. Having
said that, we perceive that the combination of the aforementioned two
factors (relieving regular routers from storing mapping table and
full RIB table) would lead to moving the mapping dissemination from
the regular BGP instance (which is used for inter-domain routing) to
a separate BGP instance only between APRs via multi-hop BGP sessions.
Though the protocol is still BGP for the ease of deployment, APRs
would run a different session (e.g., on a different TCP port) for
mapping dissemination purpose only. Other regular routers run
regular BGP instance for inter-domain routing purpose, but are
relieved from bearing the overhead of storing and propagating mapping
information or the full RIB table.
When the RIB size for most routers (other than the APRs) is reduced,
what are the prefixes that get dropped out of the RIB? Since APRs
(or ingress routers, if they are upgraded to handle caching) must
encapsulate packets towards egress routers that connect to the more
specific prefixes that have been aggregated out, the ASes must
exchange the reachability information about their own topologies, so
that routers in different ASes know how to reach each other. The
prefixes that got aggregated out of the core routing system would be
those that belong to the edge ASes. As such, Virtual Aggregation
plus mapping exchange effectively drives the overall routing system
towards the separation of edge site prefixes from the transit network
routing, a scalable routing architecture that the APT design has
depicted[eFIT_IPv6].
Again we cannot help but to point out the close resemblance between
the system we depicted above and the original APT design. On the
surface, it seems the only noticeable difference is just the names:
here we have APRs instead of APT's Default Mappers that use BGP to
exchange mapping information. But once again we must not forget an
essential difference: we reach this perceived stage towards scalable
routing through an evolutionary path, instead of requiring
installation of a new design from day one.
3.5. Step Five: Insulating the Core from Edge Churns
In the current Internet, flaps of customer prefixes are propagated to
the rest of the Internet in the form of BGP updates, i.e., routing
churns. With virtual aggregation and mapping exchange, these churns
would be reflected as mapping updates, which are disseminated through
Zhang, et al. Expires April 29, 2010 [Page 12]
Internet-Draft Scaling BGP October 2009
the interconnections of APRs. We perceive this as a benefit, as
other non-APR routers can be sheltered from updates due to edge
instabilities.
Our earlier measurement and analysis study [TopologyGrowth] has shown
that most Internet topology growth comes from the addition of
customer edge ASes. It is conceivable that as the number of customer
sites continues to increase, the amount of churns may become too much
to handle in a cost-effective way. A solution to this edge churn
problem is to insulate the mapping dissemination system from the edge
dynamics. Based on the current BGP data, our estimation shows that,
if we could remove BGP updates induced by customer prefix
instabilities, we would have reduced the total amount of routing
churns by an order of magnitude [eFIT_IPv6]. Ideally, when the link
connecting a customer site to a provider fails, the mapping system
should propagate this failure information only when the failure has a
long duration, so that every network will be aware of this failure
and choose an alternative path to reach the affected customer site.
But long lasting failures probably do not happen frequently. Short
failures, which are frequent, should not be propagated through the
mapping system. Instead, they should be handled by other means. For
example, in the APT design, the failure handling actions are data-
driven, i.e., a link failure to an edge network is not reported
unless and until there are data packets that are heading towards the
failed link. We are actively working on an evolutionary solution
that can provide equivalent data-driven handling of edge failures as
APT does.
3.6. Summary
If we can imagine a picture where all the networks in the Internet
had deployed all the steps of routing scalability improvement we
sketched above, then the Internet routing system would have converged
to a new map-encap routing architecture that resembles APT. Then
what is the fundamental difference between the evolutionary path
described in this draft and the deployment of APT?
First, we emphasize that the fundamental goal is to reduce the
routing system size, and that the separation of edges and core (or
EIDs from Locators as in LISP's terminology) itself should *not* be a
required starting point. Second, we show that the evolutionary path,
which goes through several steps with clearly identified benefits and
minimal cost at each step, can naturally converge towards the
separation as a result! We make two points from this last statement:
(1) This could also be used as an evidence that the evolution can
indeed lead to architectural changes, as it moves the system towards
the same point that a new design points to. (2) We used the phrase
"converge towards separation", rather than "achieving separation",
Zhang, et al. Expires April 29, 2010 [Page 13]
Internet-Draft Scaling BGP October 2009
because we believe that, even after a long time and many networks
have adopted the solution, it is most likely that some networks will
remain at various early stages of the evolution, some may not have
even made a single change. This is the nature of the Internet, due
to its two properties that we mentioned at the beginning: its
distributed governance, and its diversity along economical and
operational practice dimensions.
4. Evolution versus Incremental Deployability
So far our discussion has focused on a possible evolutionary path of
the routing system towards a scalable design. In this section we
would like to broaden the discussion to a more general question: Many
new designs make the claim that they are incrementally deployable.
So what is the difference between an "incrementally deployable" new
design and an evolutionary path?
We believe that one fundamental difference is that all new designs
have an implicit assumption that the entire system would eventually
move to the new design. No matter how much effort the designer puts
into the incremental deployment step of a new design, the design
itself does not start with the assumption that significant portions
of the system would never adopt it. Therefore, it is likely the case
that the assumed benefit of the new design would be achieved only
after a majority, if not the whole, of the system has deployed the
design, and that the cost of incremental deployment would be
minimized only then as well. The incremental deployment machinery is
simply to glue together the part that has made the change and the
rest that has not, to make the system function together at the
intermediate, and hopefully transient, stage. However the system as
whole would be in a sub-optimal state until the new design gets fully
deployed. LISP can serve as an example here.
In contrast, gradual evolutions in a large system depict a picture
where changes may happen here and there as needed, but there is no
expectation that the system as a whole must make a change. Whoever
adopts a step forward can gain the benefit, without waiting for
others to take actions.
The evolutionary approach recognizes that changes to the Internet can
only be a gradual process with multiple stages. At each stage,
networks that make the changes must have the incentive to do so.
More specifically,
1. Each stage focuses on an immediate problem with enough economic
impact that warrants a fix.
Zhang, et al. Expires April 29, 2010 [Page 14]
Internet-Draft Scaling BGP October 2009
2. Each stage offers a solution that solves the problem, does not
break other parts of the Internet, and can be deployed with a
reasonable cost considering the specific problem.
3. As the solution is being deployed by more and more networks, its
downside may become more pronounced and eventually requires a
fix, which leads to the next stage of the evolution.
Like many others, we too hoped that our new design, APT, could be
eventually deployed everywhere to put the routing scalability under
control. We gradually realized that it is infeasible to attempt to
roll out a new routing framework (i.e. a clean separation of edge
prefixes from the core routing system) in a vast deployed system.
The Internet Protocol, IP, was designed to accommodate heterogeneity
at subnet technology level. Today, the intrinsic heterogeneity and
distributed governance in the Internet require the accommodation of
heterogeneity at the network control plane. Solutions to routing
scalability should be control knobs on top of the deployed base to
those parties who need them, and there should not be an expectation
that the entire Internet would (eventually) move to a new design.
An evolutionary process accommodates differences at different parts
of the system, as new functions are built on top of, hence can
peacefully co-exist with, the deployed base. On the other hand, a
revolutionary new design focuses on the final outcome once the
replacement of the old by the new is done throughout the Internet.
The latter would offer a clean picture of the overall system,
assuming the final stage could be reached. The former, on the other
hand, would present a much messier or more complex picture, both
because we twist old protocols for new functions and because
different parts may do different things. As pointed out by Haldane
over 80 years ago [SizeImpact], in biological systems, "The higher
animals are not larger than the lower because they are more
complicated. They are more complicated because they are larger." We
believe the same is true for man-built systems: as the system grows
larger in size, it is necessarily becoming more complicated.
5. Acknowledgements
The authors are part of the APT team. The APT project is funded by
US National Science Foundation. The survey mentioned in Section 3.1
was conducted by Dan Jen.
6. Security Considerations
This draft is a discussion on the Internet's necessity to follow an
evolutionary path towards the future. There is no direct impact on
Zhang, et al. Expires April 29, 2010 [Page 15]
Internet-Draft Scaling BGP October 2009
the Internet security.
7. Informative References
[APT] Jen, D., Meisel, M., Massey, D., Wang, L., Zhang, B., and
L. Zhang, "APT: A Practical Tunneling Architecture for
Routing Scalability", UCLA Computer Science Departnent
Technical Report 080004, March 2008.
[APT-00] Jen, D., Meisel, M., Massey, D., Wang, L., Zhang, B., and
L. Zhang, "APT: A Practical Transit Mapping Service",
draft-jen-apt-00, July 2007.
[BGP2008] Huston, G., "BGP IN 2008 - what's changed", APRICOT
presentation, 2009, <http://apricot2009.net/
index.php?option=content&task=view&id=51>.
[FIBAggregate]
Zhang, B., Wang, L., Zhao, X., Liu, Y., and L. Zhang, "FIB
Aggregation", draft-zhang-fibaggregation-01.txt, October
2009.
[InterDomainVA]
Xu, X. and P. Francis, "Simple Tunnel Endpoint Signaling
in BGP", draft-xu-tunnel-00, February 2009.
[LISP] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis,
"Location/ID Separation Protocol (LISP)",
draft-farinacci-lisp-12, March 2009.
[NANOG44] Roisman, D., "Extending the Life of Layer 3 Switches in a
256k+ Route World", NANOG44, October 2008, <http://
www.nanog.org/meetings/nanog44/presentations/Monday/
Roisman_lightning.pdf>.
[RFC4984] Meyer, D., Zhang, L., and K. Fall, "Report from the IAB
Workshop on Routing and Addressing", RFC 4984,
September 2007.
[RRG] RRG, "IRTF Routing Research Group Home Page", <http://
tools.ietf.org/group/irtf/trac/wiki/RoutingResearchGroup>.
[SIRA] Zhang, B. and et. al., "A Secure and Scalable Internet
Routing Architecture", ACM SIGCOMM 2006 Poster Session.
[SizeImpact]
Haldane, "On Being the Right Size", 1928,
Zhang, et al. Expires April 29, 2010 [Page 16]
Internet-Draft Scaling BGP October 2009
<http://irl.cs.ucla.edu/papers/right-size.html>.
[TopologyGrowth]
Oliveira, R., Zhang, B., and L. Zhang, "Observing the
Evolution of Internet AS Topology", ACM SIGCOMM 2007.
[VA_performance]
Ballani, B., Francis, P., Jen, D., Xu, X., and L. Zhang,
"FIB Aggregation", draft-ietf-grow-va-perf-00.txt, July
2009.
[Virtual_Aggregation]
Francis, P., Xu, X., and H. Billani, "FIB Suppression with
Virtual Aggregation and Default Routes",
draft-francis-idr-intra-va-01, September 2008.
[eFIT_IPv6]
Massey, D. and et. al., "A Scalable Routing System Design
for Future Internet", ACM SIGCOMM 2007 IPv6 Workshop.
Authors' Addresses
Beichuan Zhang
Univ. of Arizona
Email: bzhang@arizona.edu
Lixia Zhang
UCLA
Email: lixia@cs.ucla.edu
Lan Wang
Univ. of Memphis
Email: lanwang@cs.memphis.edu
Zhang, et al. Expires April 29, 2010 [Page 17]