Hesham Soliman
INTERNET-DRAFT
Expires: November 2006 Qualcomm-Flarion
May, 2006
Network-based mobility considered harmful
draft-soliman-netlmm-harmful-00.txt
Status of this memo
By submitting this Internet-Draft, each author represents that any
applicable patent or other IPR claims of which he or she is aware
have been or will be disclosed, and any of which he or she becomes
aware will be disclosed, in accordance with Section 6 of 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 cite them other than as "work in progress".
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/lid-abstracts.txt
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html
This document is a submission of the IETF MIP6 WG. Comments should be
directed to the IPv6 WG mailing list, mip6@ietf.org.
Abstract
Over the last six years the IETF has developed a number of protocols
that are expected to be useful for localized mobility management
(LMM), both for IPv4 and IPv6. All of these protocols were designed
as extensions to Mobile IPv4 and Mobile IPv6. However, more recently,
the IESG has approved the formation of the NETLMM WG, which aims to
develop a network-based localized mobility management protocol. This
memo discusses the impacts of a network-based mobility management
protocol on the Internet and the IETF. The aim of this memo is to
generate discussion in this area in order to better critique the
network-based approach and show its impacts.
Soliman [Page 1]
INTERNET-DRAFT NETLMM October, 2005
Table of Contents
1. Introduction.....................................................3
2. Problems in the problem statement................................3
3. Problems with the NETLMM approach................................4
3.1. Applicability to different link layers......................5
3.2. Networks with multiple access technologies..................5
3.3. The robustness argument for end-to-end signaling............6
3.4. Mobile Vs Network Control...................................7
3.5. Network dimensioning........................................7
4. Non-technical impacts............................................8
5. Security Considerations..........................................9
6. Acknowledgements.................................................9
7. References.......................................................9
Authors' Addresses.................................................10
Soliman [Page 2]
INTERNET-DRAFT NETLMM October, 2005
1. Introduction
[MIPv6] and [MIPv4] were produced by the IETF to address mobility
management for IPv6 and IPv4, respectively. In the last 6 years, it
became evident that neither protocol is sufficient for managing
frequent mobility in a fast and efficient manner. In order to improve
the performance of Mobile IP handovers, a large effort started in the
Mobile IP WG to address what was then referred to as Localized
mobility management (LMM) for IPv4 and IPv6. In order to improve the
handover performance, two main components were addressed: 1) Reducing
the time it takes for the mobile node to detect that it has moved to
a new link, and reducing the resulting address configuration time
(specifically for IPv6); and 2) minimizing the amount of signaling
required to re-route packets to the mobile node's new location. As a
result of this effort the following specifications were produced:
[LOWLAT], [HMIPv6] and [FMIPv6]. Other supporting protocols were also
produced by the Seamoby WG: [CT] and [CARD].
Recently, the IESG has chartered the NETLMM WG to develop a network-
based approach to mobility management. The NETLMM approach excludes
the mobile node from any IP mobility decision, which is a distinct
departure from all of the mobility protocols mentioned above. This
memo critiques the approach used by the NETLMM WG and highlights some
of the problems that it can cause.
This memo begins by analyzing the problem statement that led to the
establishment of the WG then proceeds to discuss the technical
impacts of such protocol on the deployments.
2. Problems in the problem statement
The problem statement used to motivate the NETLMM effort has, in the
author's opinion, a number of factually incorrect statements. This
memo addresses the perceived problems with the current LMM approach
developed in IETF, i.e. the solutions mentioned above. The draft
lists four problems with the current approach.
The first perceived problem is that current approaches require
updates to the host software, which, however small, limit the broad
deployment of those protocols. If this concern were used as a basis
for protocol design, we would have to minimize the addition of any
new software to hosts in order to ensure wide deployment of new
features. As a result, we would maximize the dependency on new
features in the network, creating the inevitable, and sad, result of
"smart networks and dumb hosts", which is diametrically opposed to
the past and current design philosophy used in IETF. Moreover, it is
not clear why some software is acceptable in the host (e.g. the
entire TCP/IP stack, resource reservation, session control, security)
while other software is not. Finally, what proof exists today that
Soliman [Page 3]
INTERNET-DRAFT NETLMM October, 2005
relying on adding software to a host limits the deployment of new
features? The problem statement does not address these questions.
The second perceived problem with current approaches is that other
global mobility management protocols may be used in future; therefore
relying on Mobile IP extensions may not be appropriate because there
is an underlying assumption that Mobile IP will be used for global
mobility. The authors of the problem statement cite Mobike, which is
not a mobility management protocol, and HIP. Regardless of the merits
of those protocols cited as candidates for global mobility, the
argument is fundamentally flawed. The entire premise of the existing
mobility management protocols (in IPv6) is that they are completely
decoupled from the global mobility protocol. It is true that both
local and global protocols use [MIPv6], however, the current LMM
protocol [HMIPv6] does not require the use of MIPv6 for global
mobility. The same can be said for MIPv4-based approaches that use a
local HA.
The third perceived problem is that existing LMM solutions do not
support both IPv4 and IPv6. This is simply wrong. The MIP6 WG has
adopted [DSMIPv6] for this purpose and MIPv4 WG is considering an
equivalent approach [DSMIPv4].
The fourth and final perceived problem is that the current LMM
protocols require complex security mechanisms. One cannot argue with
this point since complexity, like beauty, is in the eye of the
beholder. However, it is worth noting that FMIPv6 does not yet have a
security mechanism defined, and HMIPv6 relies on IPsec, which happens
to be widely deployed and also used for MIPv6.
3. Problems with the NETLMM approach
The NETLMM approach is described on a high level in the current
charter. Simply put, NETLMM expects to place a mobility anchor point
in the visited network, which acts like a home agent. The mobile node
will be configured with an address on the anchor point's link and
keep it for the time it is located in the NETLMM domain. The anchor
point binds the mobile node's address to its current default router
(or Access Router, AR). Hence, whenever the mobile node moves, the
new AR will bind its address to the one allocated to the mobile node.
Tunneling is done between the anchor point and the AR. Therefore,
the AR's address can be seen as a Care-of address for the mobile
node. On a more detailed level, several minor changes can be made;
however, the overview above gives the general idea.
This approach raises a number of problems that were not discussed in
the process of establishing the WG. Some of these concerns are
presented in this section.
Soliman [Page 4]
INTERNET-DRAFT NETLMM October, 2005
3.1. Applicability to different link layers
The NETLMM WG charter states that the protocol will be link-layer
agnostic by running over IP. This is not completely accurate. It is
certainly true that running over IP provides some independence from
the link layer technology. However, when it comes to mobility
management, simply running over IP is not sufficient for link-layer
independence. A mobility management protocol that improves handover
performance needs to be able to adapt a sequence of events depending
on the capabilities of the link layer. For instance, Fast handovers
in IPv4 and IPv6 (FMIP) will run in different modes depending on the
underlying link layer. Broadly speaking, in the context of mobility
management, there are two types of link-layers:
1) Link-layers that allow the network to be aware of the mobile
node's next AR and anticipating its movement. Examples of such link-
layers include most cellular networks in existence today (but not
all).
2) link-layers that do not provide the knowledge of the mobile node's
next AR to the network. Examples of such link-layers include the
widely deployed WLAN technologies and some cellular link-layers.
Where the network is aware of the mobile node's next AR, the network
may be able to do the signaling on behalf of the mobile in order to
perform a predictive handover (more problems with network signaling
are discussed in the following sections). However, if the network is
not aware of the mobile node's next AR, it will not be able to
perform a predictive handover, which leaves it with reactive
handovers as the only option, which would typically result in worse
performance. Independently of the performance issue, it is important
to note that NETLMM is not link-layer agnostic when it comes to
mobility management issues.
3.2. Networks with multiple access technologies
The NETLMM problem statement document defines local mobility as
follows:
Local Mobility is mobility over a restricted area of the network
topology. Note that, although the area of network topology over
which the mobile node moves may be restricted, the actual
geographic area could be quite large, depending on the mapping
between the network topology and the wireless coverage area.
According to this definition, local mobility may take place between
two ARs connected to two different radio technologies. This poses a
serious problem for any network-based mobility management scheme. A
mobile node may wish to move some or all of its traffic to one access
technology. However, a network-based mobility management scheme
cannot be aware of the mobile node's preferences and may force one
Soliman [Page 5]
INTERNET-DRAFT NETLMM October, 2005
technology for all of the mobile node's ongoing, and possibly future,
connections. This situation can also cause operators to force a node
to be connected to a particular technology, which may not be the
preferred choice for the mobile node. This situation is not addressed
in the current charter or work done so far in the NETLMM WG. A
network-based mobility management scheme cannot handle this situation
in a reliable or deterministic manner. The flaws with a network-based
approach in this situation are:
(a) If one accepts that global mobility management is going to be
Mobile IP based then one accepts the idea that the end node should be
able to select between links to different administrative domains (or
network operators). Links to different operators can of course be of
the same or different technology. If this is a good thing, why do we
not want to provide the host with the same flexibility when different
links/technologies are available under the same local domain?
(b) Multi-homed end nodes will at some point be able to use different
links for different applications depending on link
quality/capabilities. It is easy to see that the level of complexity
increases significantly when taking into account flow movement. The
proliferation of applications and possibility that the end node is
enabled with interfaces unknown to any given network-based mobility
scheme makes this a difficult problem. How would a network based
mobility management system know which flows to move to which link?
(c) Since the coverage area of different technologies is likely to
overlap, the decision to use one technology or the other becomes a
policy decision. The end nodes will have to deal with making such
policy decisions between different networks and they should be able
to make the same decisions between different technologies. The
network operator should define metrics (like cost, loading etc) but
it should let the end host decide what to do. This is not a
philosophical point; there are concrete reasons why the host needs to
make this policy decision. For instance, the host is most
knowledgeable about the applications it runs and what radio
technologies are best suited to those applications.
3.3. The robustness argument for end-to-end signaling
End to end signaling is important and necessary in order to maintain
the end to end design philosophy of the Internet. When it comes to
localized mobility management, the end to end concept remains crucial
to the robustness of the mobility management mechanism. Handovers are
uncertain by nature and in some cases the new attachment point may
change during the handover process. This is due to the volatile
nature of the radio link at cell borders, which is typically the case
in most cellular technologies. It is also known that mobile nodes can
experience ping-pong movement, or cellular thrashing, during
handovers; i.e. the mobile node may quickly move back and forth
between two different access points for a short period of time. A
Soliman [Page 6]
INTERNET-DRAFT NETLMM October, 2005
network-based mobility management protocol can cause the mobile
node's traffic to be routed to the wrong AR, i.e. the AR that the
mobile node was expected to move to, but did not. This can result in
packet losses. In contrast, if the IP mobility signaling is initiated
from the mobile node, it would be able to discover that the next AR
has changed and inform the network of its new choice. When the action
is taken by the mobile node it can be done in a quicker manner for
predictive or reactive handovers.
3.4. Mobile Vs Network Control
One of the unwritten motivations of NETLMM is that some operators and
vendors "believe" that the network must control the handover. Lets
explore this belief a bit more. Specifically, what does network
control mean? Why is it needed? And how does a network-based mobility
management mechanism allow for more control?
One can interpret the words "Network control" to mean that the
network needs to authorize the handover of a mobile node to another
location. In fact, what is really meant by increasing "Network
control" is increasing "Operator control". There maybe reasons why
the operator needs to authorize, or at least be informed of such
move. The current LMM mechanisms do not eliminate this step. In all
of the Mobile IP-based solutions, the mobile node sends a message to
a mobility agent that responds positively or negatively to the mobile
node. Hence, it is fair to say that current LMM approaches do in fact
allow for network control of the handover. Mobile IP-based approaches
have gone further; Fast handovers for IPv4 and IPv6 allow the access
router/FA to suggest a new link for the mobile node. Hence, a
"smarter" network can suggest that a mobile node move to another
link, e.g. to better utilize its radio interface resources.
Eliminating the message from the mobile node does not increase the
network control it merely eliminates the ability of the mobile node
to request to move to a particular location.
3.5. Network dimensioning
Network dimensioning can be more difficult in the case where a
network-based approach is adopted. In a large network with millions
of mobile nodes, the network is expected to be broken down into
several local mobility domains under the same administrative domain.
Each local mobility domain would consist of several mobility anchors
(i.e. MAPs in IPv6). This is done in order to cope with the amount of
signaling and traffic generated by each mobile node. Therefore, a
form of load sharing is needed between the mobility anchor points
within each mobility domain. One of the factors involved in
dimensioning the network must be the number of users that each anchor
point can handle. This can be estimated by the amount of bandwidth
available to each anchor point, the rate of signaling it can handle,
and the number of tunnels available.
Soliman [Page 7]
INTERNET-DRAFT NETLMM October, 2005
When moving between two different mobility domains, it is strongly
recommended that the handover disruption be minimized in a similar
manner to intra mobility-domain handovers. This is especially true
for mobile nodes with ongoing sessions. Hence, it is indeed likely
that mobile nodes may continue to be attached to an anchor point in a
different mobility domain for some time after leaving that domain.
The above situation is more difficult to support when a network-based
mobility management mechanism is adopted. In particular, the
following problem arises. An anchor point may be required to setup a
security association with any access router in the network at any
time. A network administrator is suddenly forced to consider the
impacts on memory capacity and the speed of the security association
establishment at the critical handover time. This situation does not
arise when the signaling is done end-to-end because in this case only
one security association is needed, regardless of the mobile node's
location. Furthermore, the security association does not need to be
established during the critical handover time.
4. Non-technical impacts
The endorsement of more than one alternative for the same problem
needs to be strongly justified. Unfortunately this was not the case
for the NETLMM work in IETF. Both NETLMM BoFs clearly stated that the
WG will exclude the mobile node from the IP mobility management
signaling, which is not a typical requirement related to a problem,
but one designed around a solution. This premise was challenged in
both BoFs. Despite lack of clear consensus, the IESG decided to form
the WG. The problem here is two-fold: How do we decide on new WGs?
And what is the impact of solving the same problem in two different
standards? Both questions are not specific to the NETLMM WG, however,
this WG illustrates the need for a uniform and predictable process.
Developing more than one solution to solve the same problem in
different scenarios is certainly possible. However, alternatives need
to be strongly justified. In this case, the reasons are not accurate
and in most cases factually incorrect. Furthermore, the downside of
the NETLMM approach was not even discussed. This is clearly a recipe
for a bad outcome.
When justifying alternatives one needs to at least answer the
following questions to the satisfaction of the community:
- What is the problem with the current approach?
- Can this problem be solved by extending the current approach?
- What is the alternative?
- What are the pros and cons of such alternative?
As this document states, the first question was inadequately answered
and the others were not answered at all.
Soliman [Page 8]
INTERNET-DRAFT NETLMM October, 2005
5. Security Considerations
The author was using a complex, host-based, IP layer security
mechanism called IPsec while writing this draft.
6. Acknowledgements
A lot of the points made in this document were discussed with,
inspired by, or produced during discussions with (in alphabetical
order): Scott Corson, Vince Park, Geroge Tsirtsis. Their input has
significantly improved the quality of this document.
7. References
[CARD] CARD Design team, Liebsch, M., Singh, A., Chaskar H. and D.
Funato, "Candidate Access Router Discovery", RFC 4066 March
2003.
[CT] Loughney, J. Ed., Nakhjiri, M., Perkins, C. and R. Koodli,
"Context Transfer Protocol (CXTP)", RFC 4067 July 2005.
[DSMIPv4] Tsirtsis, G., Soliman, H., and V. Park, " Dual Stack Mobile
IPv4 ", draft-tsirtsis-v4v6-mipv4-01/
[DSMIPv6] H. Soliman et al, " Mobile IPv6 for Dual Stack Hosts and
Routers (DSMIPv6)", draft-ietf-mip6-nemo-v4traversal-01.
[HMIPv6] Soliman, H., Castelluccia, C., ElMalki, K., and L. Bellier,
"Hierarchical Mobile IPv6 Mobility Management (HMIPv6)",
RFC 4140, August 2005.
[IPv6] S. Deering and B. Hinden, "Internet Protocol version 6 (IPv6)
specification", RFC 2460
[KEYWORDS] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997.
[LOWLAT] K. ElMalki, Ed., "Low Latency handovers for Mobile IP"
[MIP-PB] Tsirtsis, G., and H. Soliman, "Mobility management for
Dual stack mobile nodes, A Problem Statement",
draft-ietf-mip6-dsmip-problem-01.txt, July 2005.
[MIPv4] C. Perkins, "Mobility Support for IPv4", RFC3344
[MIPv6] D. Johnson, C. Perkins and J. Arkko, "Mobility Support in
IPv6", RFC 3775, June 2004.
[NEMO] Devarapalli, V., Wakikawa, R., Petrescu, A., and P. Thubert,
"Network Mobility (NEMO) Basic Support protocol", RFC 3963,
January 2005.
Soliman [Page 9]
INTERNET-DRAFT NETLMM October, 2005
[SEC] Arkko, J., Devarapalli, V., and F. Dupont, "Using IPsec to
Protoect Mobile IPv6 Signaling between Mobile Nodes and Home
Agents", RFC 3776, June 2004.
[SNRIO] Larsson, T., Gustafsson, E., and H. Levkowetz, "Use of MIPv6
in IPv4 and MIPv4 in IPv6 networks", draft-larsson-v6ops-
mip-scenarios-01.txt, February 2004.
Authors' Addresses
Hesham Soliman
Qualcomm-Flarion Technologies
E-mail: Hesham@Qualcomm.com
Intellectual Property Statement
The IETF takes no position regarding the validity or scope of any
Intellectual Property Rights or other rights that might be claimed to
pertain to the implementation or use of the technology described in
this document or the extent to which any license under such rights
might or might not be available; nor does it represent that it has
made any independent effort to identify any such rights. Information
on the IETF's procedures with respect to rights in IETF Documents can
be found in RFC 3667 (BCP 78) and RFC 3668 (BCP 79).
Copies of IPR disclosures made to the IETF Secretariat and any
assurances of licenses to be made available, or the result of an
attempt made to obtain a general license or permission for the use of
such proprietary rights by implementers or users of this
specification can be obtained from the IETF on-line IPR repository at
http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights that may cover technology that may be required to implement
this standard. Please address the information to the IETF at ietf-
ipr@ietf.org.
Full Copyright Statement
Copyright (C) The Internet Society (2006). This document is subject
to the rights, licenses and restrictions contained in BCP 78, and
except as set forth therein, the authors retain all their rights.
Disclaimer of Validity
Soliman [Page 10]
INTERNET-DRAFT NETLMM October, 2005
This document and the information contained herein are provided on an
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
This Internet-Draft expires September, 2006.
Soliman [Page 11]