Internet Research Task Force Patrick Frejborg
Internet Draft September 16, 2010
Intended status: Experimental
Expires: March 2011
Hierarchical IPv4 Framework
draft-frejborg-hipv4-09.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 March 16, 2011.
Copyright Notice
Copyright (c) 2010 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
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
Frejborg
Internet-Draft Hierarchical IPv4 Framework September 2010
Abstract
This document describes a framework how the current IPv4 address
space can be divided in two new address categories; a core address
space (Area Locators, ALOC) that is globally unique and an edge
address space (Endpoint Locators, ELOC) that is regionally unique -
in the future the ELOC space is only significant in a private network
or in a service provider domain, therefore a 32 x 32 bit addressing
scheme is possible. The core/edge address grouping will enable a new
level of hierarchy in the routing architecture of Internet. The
hierarchical IPv4 framework is backwards compatible with the current
IPv4 framework; it will also discuss a method to decouple the
location and identifier functions, future applications can make use
of the separation. The framework requires extensions to the existing
Domain Name System, the existing IPv4 stack of the endpoints,
security nodes and to routers in the Internet. The framework can be
implemented incrementally to endpoints, DNS, security nodes and
routers. This document is a product within the IRTF Routing Research
Group. The views in this document are considered controversial by the
IRTF Routing Research Group (RRG) but the group reached a consensus
that the document should still be published.
Table of Contents
1. Requirements Notation........................................3
2. Introduction.................................................3
3. Definitions of terms.........................................6
4. An overview of the hIPv4 framework...........................7
5. Mandatory extensions to current architectures (unicast).....10
6. The header of a hIPv4 packet................................11
7. ALOC use cases..............................................15
8. Life of a hIPv4 session.....................................17
9. Overlapping Source and Destination ELOC prefixes/ports......20
10. Traceroute considerations..................................21
11. Multicast considerations...................................21
12. Traffic engineering considerations.........................22
13. Large encapsulated packets.................................24
14. Mobility considerations....................................24
15. Affected Applications and Implications.....................26
16. The Future Role of the LSR.................................27
17. Transition considerations..................................29
18. Security Considerations....................................29
19. IANA Considerations........................................30
20. Conclusion.................................................30
21. References.................................................31
21.1. References............................................31
21.2. Informative References................................31
Frejborg Expires March 16, 2011 [Page 2]
Internet-Draft Hierarchical IPv4 Framework September 2010
22. Acknowledgements...........................................33
Appendix A. Short term & future IPv4 address allocation policy.34
Appendix B. Multi-homing becomes multi-pathing.................36
Appendix C. Transition Arguments...............................40
Appendix D. Integration with CES architectures.................42
1. Requirements Notation
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in [RFC2119].
2. Introduction
The hierarchical IPv4 framework (hIPv4) has been developed to address
the issues discussed in the [IAB report] from the Routing and
Addressing Workshop that was held by the Internet Architecture Board
(IAB) on October 18-19, 2006, in Amsterdam, Netherlands. Note! The
hierarchical IPv4 framework is abbreviated as hIPv4 which is close to
the abbreviation of Host Identity Protocol [HIP], thus the reader
need to pay attention to these two abbreviations - hiPV4 and HIP are
two different architectures.
The current addressing (IPv4) and the future addressing (IPv6)
schemes of Internet are single dimensional by their nature. This
limitation, i.e. the single level addressing scheme, has created some
roadblocks for further growth of Internet. If we compare Internet's
current addressing schemes to other global addressing or location
schemes we can notice that the other schemes use several levels in
their structures. E.g. the postal system uses street address, city
and country to locate a destination. Also to locate a geographical
site we are using longitude and latitude in the cartography system.
The other global network, the Public Switched Telephone Network
(PSTN), have been build upon a three level numbering scheme that have
enabled a hierarchical signaling architecture. By expanding the
current IPv4 addressing scheme from a single level to a two level
addressing structure most of the issues discussed in the IAB report
can be solved. Also a hierarchical addressing scheme would better
describe the Internet we have in place today - it seems that the
architecture of Internet changed quite radically from the designed
architecture by the introduction of [RFC1918] which divides the hosts
into three categories and the address space into two categories -
i.e. globally unique and private address spaces. This invention
provided further growth of Internet and extended the life of the IPv4
address space, much longer than expected. RFC1918 didn't solve the
multi-homing requirements for category 3 hosts - multi-homing with
globally unique addresses has imposed some challenges for the routing
Frejborg Expires March 16, 2011 [Page 3]
Internet-Draft Hierarchical IPv4 Framework September 2010
architecture that [RRG] is addressing in the [RRG Recommendation]
report. Almost all proposals in the report suggest a core and edge
locator separation or elimination to create a scalable routing
architecture. The core locator space can be viewed to be similar as
the globally unique address space and the edge locator space similar
as the private address space. RFC1918 has already demonstrated that
Internet scales better with the help of categorized address spaces,
the RRG proposals are also indicating that Internet will scale even
further by introducing core and edge locators - then why not then
change the addressing scheme (both IPv4 and IPv6 addressing schemes
though this document is only focusing on IPv4) to better reflect the
current and forthcoming Internet routing architecture? Continuing the
use of a flat addressing scheme combined with core(global) and
edge(private) locator(address) categories, the routing architecture
must implement additional mechanisms - such as NAT, tunneling or
locator rewriting with the help of an identifier - to overcome the
mismatch, and information is lost or hided for the endpoints. With a
two-level addressing scheme these additional mechanisms can be
removed and core/edge locators can be used to create new routing and
forwarding directives.
A convenient way to understand the two-level addressing scheme of the
hIPv4 framework is to compare it to the PSTN numbering scheme (E.164)
which uses country codes, national destination codes and subscriber
numbers. The Area Locator (ALOC) prefix in the hIPv4 addressing
scheme can be considered similar as to the country code in PSTN, i.e.
the ALOC prefix locates an area in Internet, the area is called an
ALOC realm. The Endpoint Locator (ELOC) prefixes in hIPv4 can be
compared to the subscriber numbers in PSTN, i.e. the ELOC is
regionally unique (in the future, locally unique) at the attached
ALOC realm - the ELOC can also be attached simultaneously to several
ALOCs realms.
By inserting the ALOC and ELOC elements as a shim header (similar as
in [MPLS] and [RBridge] architectures) between the IPv4 header and
the transport protocol header, a hIPv4 header is created. From the
network point of view, the hIPv4 header "looks and feels like" an
IPv4 header - thus fulfilling some of the goals as outlined in [EIP]
and in the early definition of [Nimrod] - outcome is that the current
forwarding plane do not need to be upgraded though some minor changes
are needed in the control plane (e.g. ICMP extensions).
Another important influence source are the report and presentations
from the [Dagstuhl] workshop that declared "a future Internet
architecture must hence decouple the functions of IP addresses as
names, locators, and forwarding directives in order to facilitate the
growth and new network-topological dynamisms of the Internet".
Frejborg Expires March 16, 2011 [Page 4]
Internet-Draft Hierarchical IPv4 Framework September 2010
Therefore, identifier elements need to be added to the hIPv4
framework to provide a path for future applications to be able to
remove the current dependency of the underlying network layer
addressing scheme (local and remote IP address tuples). Multipath
transport protocols, such as [SCTP] and the currently under
development Multipath TCP [MPTCP], are the most interesting
candidates to enable an identifier functionality for the hIPv4
framework. Especially MPTCP is interesting from hIPv4's point of view
- one of the main goals of MPTCP is to provide backwards
compatibility with current implementations, hIPv4 share the same
goal. MPTCP itself do not provide a host identifier solution as e.g.
Host Identity Protocol [HIP] does, instead MPTCP is proposing a token
- with local meaning - to manage and bundle subflows under one
session between two endpoints. The token can be considered to have
the characteristics of a session identifier, providing a generic
cookie mechanism for the application layer and creating a session
layer between the application and transport layer. Thus the usage of
a session identifier will provide a mechanism to improve mobility,
both in site and endpoint mobility scenarios.
Since the session identifier is improving site and endpoint mobility,
routing scalability is improved by introducing a hierarchical address
space, why then add a host identifier solution to the hIPv4
framework? Introducing a host identifier scheme, as described in
[HIP], Identifier/Locator Network Protocol [ILNP] and Name-Based
Sockets [NBS], might ease or remove the locator renumbering
dependencies at security nodes (firewalls) that are used to scope
security zones but this approach would change fundamentally the
current deployed security architecture. However, combining a host
identifier scheme with [DNSSEC] is interesting. Today security zones
are scoped by using locator prefixes in the security rule sets -
instead FQDN could be used in the rule sets and the renumbering of
locator prefixes would no longer be depended upon the security rule
sets in the security nodes. Another interesting aspect is that a FQDN
is and needs to be globally unique, the ALOC prefix must be globally
unique but ELOC prefixes are only regionally unique, in long term
only locally unique. Nevertheless, combining host identifiers with a
security architecture and [DNSSEC] needs further studies.
In order to provide multi-homing and mobility capabilities for
single-path transport protocols, e.g. TCP and UDP a host identifier
scheme is needed. The host identifier scheme can also be used to
create a bidirectional NAT traversal solution with a locator
translation map consisting of private locator prefixes and public
host identities at the border router.
Frejborg Expires March 16, 2011 [Page 5]
Internet-Draft Hierarchical IPv4 Framework September 2010
3. Definitions of terms
Regional Internet Registry (RIR):
Is an organization overseeing the allocation and registration of
Internet number resources within a particular region of the world.
Resources include IP addresses (both IPv4 and IPv6) and autonomous
system numbers.
Locator:
A locator is a name for a point of attachment within the topology
at a given layer. Objects that change their point of attachment(s)
will need to change their associated locator(s). In the hIPv4
framework two types of locators have been defined, Area Locator
(ALOC) and Endpoint Locator (ELOC).
Global Locator Block (GLB):
An IPv4 address block that is globally unique.
Area Locator (ALOC):
An IPv4 address (/32) assigned to locate an ALOC realm in
Internet. The ALOC is assigned by a RIR to a service provider; the
ALOC is globally unique because it is allocated from the GLB.
Endpoint Locator (ELOC):
An IPv4 address assigned to locate an endpoint in a local network.
The ELOC block is assigned by a RIR to a service provider or to an
enterprise. The ELOC block is only unique in a geographical
region. The final policy of uniqueness shall be defined by the
RIRs.
ALOC realm:
An area in the Internet with at least one attached Locator Swap
Router (LSR), also an ALOC must be assigned to the ALOC realm. The
RIB of an ALOC realm holds both local ELOC prefixes and global
ALOC prefixes. An ALOC realm exchanges only ALOC prefixes with
other ALOC realms.
Locator Swap Router (LSR):
A router or node which is capable to process the hIPv4 header;
once the header is processed the LSR will forward the packet upon
Frejborg Expires March 16, 2011 [Page 6]
Internet-Draft Hierarchical IPv4 Framework September 2010
the destination address in the IP header. The LSR must have the
ALOC assigned as its locator.
Locator Header:
A 4 or 12 byte field, inserted between the IP header and transport
protocol header. If a host identifier scheme is used the size of
the locator header is further expanded.
Identifier:
An identifier is the name of an object at a given layer;
identifiers have no topological sensitivity, and do not have to
change, even if the object changes its point(s) of attachment
within the network topology.
Session:
A session is a semi-permanent interactive information exchange
between endpoints that is established at a certain time and torn
down at a later time.
Provider Independent Address Space (PI addresses):
An IPv4 address block which is assigned by a Regional Internet
Registries directly to an end user organization.
Provider Aggregatable Address Space (PA addresses):
An IPv4 address block assigned by a Regional Internet Registry to
an Internet Service Provider which can be aggregated into a single
route advertisement.
4. An overview of the hIPv4 framework
In this section we will discuss the roles of the new elements,
introduced by the hIPv4 framework, and their dependencies.
As mentioned in the introduction section the role of an Area LOCator
(ALOC) prefix is similar as to a country code in PSTN. I.e. the ALOC
prefix provides a location functionality of an area within an
Autonomous System (AS), or an area spanning over a group of AS, in
Internet. An AS can have several ALOC prefixes assigned, e.g. due to
traffic engineering requirements. The ALOC prefix will be used for
routing and forwarding purposes in Internet, thus the ALOC prefix
must be globally unique and is allocated from an IPv4 address block.
Frejborg Expires March 16, 2011 [Page 7]
Internet-Draft Hierarchical IPv4 Framework September 2010
This globally unique IPv4 address block is called the Global Locator
Block (GLB).
When an area within an AS (or a group of AS) are assigned an ALOC
prefix the area has the potential to become an ALOC realm. In order
to establish an ALOC realm more elements, further than the ALOC
prefix, are needed. One or multiple Locator Swap Routers (LSR) must
be attached to the ALOC realm. A LSR element is a node capable of
swapping the prefixes of the IP header and the new shim header,
called the locator header. The swap mechanism of the headers is
described in detail in section 8, step 4. Today's routers do not
support the LSR functionality. Therefore the new functionality will
most likely be developed on an external device attached to a router
belonging to the ALOC realm. The external LSR might be a computer
with two interface attached to a router, the first interface
configured with the prefix of the ALOC and the second interface with
any IPv4 prefix. The LSR do not make us of dynamic routing protocols,
neither a forwarding information base (FIB) nor a cache is needed -
the LSR is producing a service, i.e. swapping headers. The swap
mechanism is applied on per packet basis and the information needed
to carry out the swap is included in the locator header of the hIPv4
packet. Thus a computer with enough computing and I/O resources is
sufficient to take the role as a LSR. Later on, the LSR functionality
might be integrated into the forwarding plane of a router. One LSR
can not handle all the incoming traffic designated for an ALOC realm;
it would also create a potential single point of failure in the
network. Therefore, several LSRs might be installed in the ALOC realm
and the LSRs shall use the ALOC prefix as their locator and the
routers are announcing the ALOC prefix as an anycast address within
the local ALOC realm. Also, the ALOC prefix is advertised throughout
the DFZ by BGP mechanisms. The placement of the LSRs in the network
will influence on the ingress traffic to the ALOC realm.
Since the forwarding paradigm of multicast packets is quite different
from forwarding unicast packets the multicast functionality will have
an impact on the LSR. Also, the multicast LSR (mLSR) functionality is
not available on today's routers, an external device is needed -
later on the functionality might be integrated to the routers. The
mLSR shall take the role of an anycast RP with MSDP and PIM
capabilities, but to forward packets a FIB is not required. As with
the LSR, the multicast hIPv4 packets are carrying all needed
information in their headers in order to apply the swap, for details
see section 11.
The ALOC realm is not yet fully constructed, we can now locate the
ALOC realm in Internet but to locate the endpoints attached to the
ALOC realm a new element is needed, i.e. the Endpoint LOCator (ELOC).
Frejborg Expires March 16, 2011 [Page 8]
Internet-Draft Hierarchical IPv4 Framework September 2010
As mentioned in the introduction section the ELOC prefixes can be
considered similar as to the subscriber numbers in PSTN. Actually,
the ELOC is not a new element; the ELOC is a redefinition of the
current IPv4 address configured at an endpoint. The redefinition is
applied because when the hIPv4 framework is fully implemented the
global uniqueness of the IPv4 addresses are no longer valid and a
more regional address allocation policy of IPv4 addresses can be
deployed as discussed in appendix A. The ELOC prefix will only be
used for routing and forwarding purposes inside the local and remote
ALOC realms, the ELOC prefix is not used in the intermediate ALOC
realms. When an initiator is establishing a session to a responder
residing outside the local ALOC realm the destination address field
in the IP header of an outgoing packet is no longer the remote
destination address (ELOC prefix) - instead the remote ALOC prefix is
installed in the destination address field of the IP header. Because
the destination address is an ALOC prefix, the intermediate ALOC
realms do not need to carry the ELOC prefixes of other ALOC realms in
their RIB - it is sufficient for the intermediate ALOC realms to
carry only the ALOC prefixes. Outcome is that the RIB and FIB tables
at each ALOC realm will be reduced when the hIPV4 framework is fully
implemented. The ALOC prefixes are still globally unique and must be
installed in the DFZ - thus the service provider can't control the
growth of the ALOC prefixes but she/he can control the amount of
local ELOC prefixes in her/his local ALOC realm.
When the hIPv4 packet arrives at the remote ALOC realm it will be
forwarded to the nearest LSR, since the destination address is the
remote ALOC prefix. When the LSR has swapped the hIPv4 header, the
destination address is the remote ELOC, thus the hIPv4 packet will be
forwarded to the final destination at the remote ALOC realm. An
endpoint using an ELOC prefix can be attached simultaneously to two
different ALOC realms without the requirement to deploy a classical
multi-homing solution, for details see section 14.
Next, how can we locate the remote ELOC (endpoint) and remote ALOC
realm in Internet, also how to assemble the header of the hIPv4
packet? Another matter is that the addressing structure is no longer
single dimensional; instead a second level has been added on top of
the old one. It is obvious that the Domain Name System needs to
support a new record type so that the ALOC information can be
distributed to the endpoints. To construct the header of the hIPv4
packet either the endpoint or an intermediate node (e.g. a proxy)
should be used. A proxy solution is complicated, the proxy needs to
listen to DNS messages and a cache solution does have scalability
issues.
Frejborg Expires March 16, 2011 [Page 9]
Internet-Draft Hierarchical IPv4 Framework September 2010
A better solution is to extend the current IPv4 stack at the
endpoints so that the ALOC and ELOC elements are incorporated at the
endpoint's stack, but backwards compatibility must be preserved. Most
of the application will not be aware of the extensions - other
applications, such as Mobile IP, SIP, IPsec AH and so on (see section
15) will suffer and can not be used outside their ALOC realm when the
hIPv4 framework is fully implemented - unless the applications are
upgraded. The reason is that these applications are depending upon
the underlying network addressing structure to e.g. identify an
endpoint.
5. Mandatory extensions to current architectures (unicast)
To implement the hierarchical IPv4 framework some basic rules are
needed:
1. The DNS architecture must support a new extension, i.e. an A type
Resource Record should be able to carry ALOC prefix(es).
2. The hIPv4 capable endpoint shall have information about the local
ALOC prefix(es); the local ALOC prefix(es) can be configured
manually or provided via a new DHCP option.
3. A globally unique IPv4 address block shall be reserved; this block
is called the Global Locator Block (GLB). A service provider can
have one or several ALOC prefixes allocated from the GLB.
4. ALOC prefixes are announced via current BGP protocol to adjacent
service providers and multi-homed enterprises, the ALOC prefixes
are installed in the RIB of the DFZ. When the hIPV4 framework is
fully implemented only ALOC prefixes are announced between the
service providers and multi-homed enterprises.
5. A hIPv4 capable ALOC realm must have one or several LSRs attached
to its realm. The ALOC prefix is configured as an anycast IP
address on the LSR. The anycast IP address is installed to
appropriate routing protocols in order to be distributed to the
DFZ.
Frejborg Expires March 16, 2011 [Page 10]
Internet-Draft Hierarchical IPv4 Framework September 2010
6. The IPv4 socket API at endpoints must be extended to support local
and remote ALOC prefixes. The modified IPv4 socket API must be
backwards compatible with the current IPv4 socket API. The
outgoing hIPv4 packet must be assembled by the hIPv4 stack with
the local IP address from the socket as the source address and the
remote ALOC prefix as the destination address in the IP header.
The local ALOC prefix is inserted in the ALOC field of the locator
header. The remote IP address from the socket API is inserted in
the ELOC field of the locator header.
6. The header of a hIPv4 packet
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|Version| IHL |Type of Service| Total Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Identification |Flags| Fragment Offset |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Time to Live | Protocol | Header Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Source Address |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Destination Address |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Options | Padding |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|A|I|S| FI|VLB|L| Protocol | LH Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Area Locator (optional) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Endpoint Locator (optional) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| LH Length (optional) | Padding (optional) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Version: 4 bits
The Version field is identical to that of RFC 791.
IHL: 4 bits
Internet Header Length field is identical to that of RFC 791.
Type of Service: 8 bits
Frejborg Expires March 16, 2011 [Page 11]
Internet-Draft Hierarchical IPv4 Framework September 2010
The Type of Service is identical to that of RFC 791.
Total Length: 16 bits
The Total Length field is identical to that of RFC 791.
Identification: 16 bits
The Identification field is identical to that of RFC 791.
Flags: 3 bits
The Flags field is identical to that of RFC 791.
Fragment Offset: 13 bits
The Fragment Offset field is identical to that of RFC 791.
Time to Live: 8 bits
The Time to Live field is identical to that of RFC 791.
Protocol: 8 bits
A new protocol number must be assigned for hIPv4.
Header Checksum: 16 bits
The Header Checksum field is identical to that of RFC 791.
Source Address: 32 bits
The Source Address field is identical to that of RFC 791.
Destination Address: 32 bits
The Destination Address field is identical to that of RFC 791.
Options and Padding: Variable length
The Options and padding field is identical to that of RFC 791.
ALOC Realm Bit, A-bit: 1 bit
When the source and destination endpoints reside in different ALOC
realms, the A-bit is set to 1 and the Area and Endpoint Locator
fields must be used in the locator header. The size of the locator
Frejborg Expires March 16, 2011 [Page 12]
Internet-Draft Hierarchical IPv4 Framework September 2010
header is 12 bytes. When the A-bit is set to 0 the source and
destination endpoints reside within the same ALOC realm, the Area
and Endpoint Locator shall not be used in the locator header. The
size of the locator header is 4 bytes.
Identifier Bit, I-bit: 1 bit
The identifier bit is set to 1 if the endpoint is using a host
identifier scheme within the locator header. The host identifier
scheme must indicate how much the size of the locator header is
increased, thus the Locator Header Length field is also added to
the locator header.
Swap Bit, S-bit: 1 bit
The initiator sets the swap bit to 0 of the hIPv4 packet. A LSR
will set this bit to 1 when it is swapping the source and
destination addresses of the IP header with the ALOC and ELOC
prefixes of the locator header.
Forwarding Indicator, FI-bits: 2 bits
The purpose of Forwarding Indicator field is to provide a mechanism
for future forwarding plane to identify which forwarding
information base should be used for inter-ALOC realm sessions. The
new forwarding plane will remove the swap functionality of IP and
locator header values for both unicast and multicast sessions.
Outcome is that the IP and transport protocol headers will remain
intact, only FI and Locator Header Checksum values in the locator
header will alternate.
00: Local ALOC exit forwarding mode. The initiator shall set the
FI-bits to 00 and the ALOC prefix in the locator header is used to
forward the packets to the border router which is the owner of the
local ALOC prefix. The border router shall change the FI-bits to
01.
01: DFZ forwarding mode. The local ALOC border router shall
forward the packets upon the destination address in the IP header.
The DFZ routers shall forward the packets based upon the
destination address unless the destination address matches the
local ALOC prefix - when this situation occurs the packet enters
the remote ALOC realm and the remote border router shall change
the FI-bits to 11.
Frejborg Expires March 16, 2011 [Page 13]
Internet-Draft Hierarchical IPv4 Framework September 2010
11: Remote ALOC entry forwarding mode. The remote ALOC border
router and following routers shall forward the packets upon the
ELOC prefix in the locator header.
10: Inter-ALOC RFP check mode. The local ALOC border router change
the FI-bits to 10 and following inter-ALOC routers on the shared
tree shall apply RFP check upon the ALOC prefix in the locator
header.
Valiant Load-Balancing, VLB-bits: 2 bits (optional, subject for
further research)
The purpose of Valiant Load-Balancing field is to provide a
mechanism for multipath enabled transport protocols to request
explicit paths in the network for subflows, which are component
parts of a session between two endpoints. The subflow path request
can be set as following:
00: Latency sensitive application, only one single subflow (i.e.
multipath not applied), shortest path through the network is
requested.
01: First subflow, shortest path or Valiant Load-Balancing might
be applied.
11: Next subflow(s), Valiant Load-Balancing should be applied
Load-Balanced, L-bit: 1 bit (optional, subject for further research)
The initiating endpoint must set the L-bit to zero. A Valiant Load-
Balancing capable node can apply VLB switching for the session if
the value is set to zero; if the value is set to 1 VLB switching is
not allowed. When VLB switching is applied for the session the node
must set the value to 1.
Protocol: 8 bits
The Protocol field is identical to that of RFC 791
Locator Header Checksum: 16 bits
A checksum is calculated on the locator header only. The checksum
is computed at the initiator, recomputed at the LSR and verified at
the responder. The checksum algorithm is identical to that of RFC
791.
Area Locator (optional): 32 bits
Frejborg Expires March 16, 2011 [Page 14]
Internet-Draft Hierarchical IPv4 Framework September 2010
An IPv4 address, the ALOC is assigned by a RIR to a service
provider. The ALOC is globally unique because it is allocated from
the GLB.
Endpoint Locator (optional): 32 bits
An IPv4 address, the ELOC block is assigned by a RIR to a service
provider or to an enterprise. The ELOC block is only unique in a
geographical region, if forwarding plane is replaced the ELOC block
needs only to be locally significant.
Locator Header Length (optional): 16 bits
Locator Header Length is the length of the locator header. Locator
Header Length is applied when the Identifier bit is set to 1. Host
identifier parameters are inserted to the locator header after this
field.
Padding (optional): variable
The locator header padding is used to ensure that the locator
header ends on a 32 bit boundary. The padding is zero.
7. ALOC use cases
Since ALOC is the main component that has been added to the current
IPv4 framework and enabled the hIPv4 framework several ALOC use cases
are explored in this section. As mentioned in previous sections ALOC
is describing an area in Internet, the area can span over several
Autonomous Systems (AS) or if the area is equal to an AS you can say
that the ALOC is describing an AS. When the ALOC is describing an
area it is hereafter called an anycast ALOC. The ALOC can also be
used to describe a specific node between two ALOC realms, e.g. a node
installed between a private and an ISP ALOC realm or between two
private ALOC realms. In this use case the ALOC is describing an
attachment point, e.g. where a private network is attached to
Internet - the ALOC type is hereafter called a unicast ALOC. The main
difference between anycast and unicast ALOC types is that in anycast
ALOC scenarios ELOC routing information is shared between the
attached ALOC realms. In a unicast ALOC scenario no ELOC routing
information is shared between the attached ALOC realms. Unicast ALOC
functionalities shall not yet be deployed between private and ISP
ALOC realms - it would require too many locators from the GLB space -
instead unicast ALOC functionality shall be used to separate private
ALOC realms.
Frejborg Expires March 16, 2011 [Page 15]
Internet-Draft Hierarchical IPv4 Framework September 2010
ALOC space is divided into two types, a globally unique ALOC space
(GLB) that is installed in DFZ and private ALOC space that is used
inside private networks. Private ALOC are using the same locator
space as defined in [RFC1918], the private ALOC must be unique inside
the private network and should not overlap with private ELOC
prefixes. Only ISP should be allowed to apply for global ALOC
prefixes, for further discussions see appendix A. The ISP should
aggregate global ALOC prefixes as much as possible in order to reduce
the size of the routing table in DFZ.
When a user logs on the enterprise's network the endpoint will
receive via DHCP (or manually configured) the following locator
prefixes:
o one ELOC prefix for each network interface
o one private ALOC prefix if e.g. the enterprise network spans
over a small region or
o several private ALOC prefixes if e.g. the enterprise network
spans over several regions (i.e. uses long distance
connections) or recently have been merged with another
enterprise
o one or several global ALOC prefixes, these ALOC describes how
the enterprise network is attached to Internet
As the user establishes a session to a remote endpoint DNS is usually
used to resolve remote locator prefixes, DNS will return ELOC and
ALOC prefixes of the remote endpoint. If no ALOC prefixes are
returned a classical IPv4 session is initiated to the remote
endpoint. When ALOC prefixes are returned the initiating endpoint is
comparing the ALOC prefixes with its own local ALOC prefixes (that
are provided via DHCP or manually configured):
o if the remote ALOC prefix is from the private ALOC space the
initiator shall use the given private ALOC prefix for the
session.
Two use cases exist to design a network to use private ALOC
functionality. The remote endpoint is far away and in order to
improve performance for the session a multipath transport protocol
should be used. The other use case is when the remote endpoint
resides in a network that recently has been merged and private ELOC
[RFC1918] spaces overlaps if no renumbering is applied. One or
several node ALOC solutions are needed in the network between the
initiator and responder. For long distance sessions with no
Frejborg Expires March 16, 2011 [Page 16]
Internet-Draft Hierarchical IPv4 Framework September 2010
overlapping ELOC prefixes, anycast or unicast ALOC solutions can be
deployed. Third use case follows, again the initiator compares
returned ALOC prefixes from DNS with own local ALOC prefixes:
o if the remote ALOC prefix is from the global ALOC space and
the remote ALOC doesn't match the given global ALOC prefix the
initiator shall use the given global ALOC prefix for the
session.
In this use case the remote endpoint resides outside the enterprise's
own private network, the global remote ALOC prefixes indicates how
the remote network is attached to Internet. When a multipath
transport protocol is used the subflows can be routed via separate
border routers to the remote endpoint - both at the local and remote
sites, if both are multi-homed. The egress hIPv4 packet (the
initiator's packets in the local network) can be identified by the
protocol value in the IP header, routed to an explicit path (e.g.
MPLS LSP, L2TPv3 tunnel etc) upon the ALOC prefix in the locator
header - a local ALOC exit routing scheme can be designed, for
further details, see section 12 and appendix B.
Fourth use case is to leverage the private and global ALOC
functionalities to be aligned with the design and implementation of
split-DNS solutions.
The fifth use case is for residential users, a residential user may
use one or several ALOC prefixes; it depends upon the service offer
and network design of the ISP. If the ISP prefers to offer advanced
support for multipath transport protocols and local ALOC exit
routing, the residential user is provided with several ALOC prefixes.
The ALOC provided for residential users is taken from the GLB space
and anycast ALOC functionality is applied.
8. Life of a hIPv4 session
This section provides an example of a hIPv4 session between two hIPv4
endpoints; an initiator and a responder residing in different ALOC
realms.
When the hIPv4 stack is assembling the packet for transport the hIPv4
stack shall decide if a classical IPv4 or a hIPv4 header is used upon
the ALOC information received by a DNS reply. If the initiator's
local ALOC prefix equals the responder's ALOC prefix there is no need
to use the hIPv4 header for routing purposes, because both the
initiator and responder reside in the local ALOC realm. The packet is
routed upon the IP header since the packet will not exit the local
ALOC realm. When the local ALOC prefix doesn't match the remote ALOC
Frejborg Expires March 16, 2011 [Page 17]
Internet-Draft Hierarchical IPv4 Framework September 2010
prefix a hIPv4 header must be assembled because the packet needs to
be routed to a remote ALOC realm.
A session between two endpoints inside an ALOC realm might use the
locator header - not for routing purposes, but to make use of Valiant
Load-Balancing [VLB] for multipath enabled transport protocols or
make use of a host identifier scheme. The initiator can add the
locator header to the packet and by setting the VLB bits to 01
indicating to the responder and intermediate routers that VLB is
requested for the subflow. Because this is an intra-ALOC realm
session there is no need to add ALOC and ELOC fields to the locator
header, thus the size of the locator header will be 4 bytes. If a
host identifier scheme is applied for the session (intra-ALOC or
inter-ALOC), the scheme must set the I-bit to 1 and make use of the
Locator Header Length field. Host identifier information is inserted
to the locator header after the Locator Header Length field.
How a hIPV4 session is established follows:
1. The initiator queries the DNS node; the hIPv4 stack notice that
the local and remote ALOC doesn't match and therefore must use the
hIPv4 header for the session. The hIPv4 stack of the initiator
must assemble the packet in the following way:
a. set local IP address from API in the source address field of
the IP header
b. set remote IP address from API in the ELOC field of the
locator header
c. set local ALOC prefix in the ALOC field of the locator header
d. set remote ALOC prefix in the destination address field of the
IP header
e. set the transport protocol value in the protocol field of the
locator header and set the hIPv4 protocol value in the
protocol field of the IP header
f. set the desired parameters in the A-, I-, S-, FI-, VLB-, and
L-fields of the locator header
g. calculate IP-, locator- and transport protocol header
checksums, transport protocol header calculation do not
include the locator header fields. When completed the packet
is transmitted.
Frejborg Expires March 16, 2011 [Page 18]
Internet-Draft Hierarchical IPv4 Framework September 2010
2. The hIPv4 packet is routed throughout Internet upon the
destination address of the IP header.
3. The hIPv4 packet will reach the closest LSR of the remote ALOC
realm. When the LSR notice that the packet matches the local ALOC
prefix the LSR must:
a. verify the received packet that it uses the hIPv4 protocol
value in the protocol field of the IP header
b. verify IP-, locator- and transport protocol header checksums,
transport protocol header verification do not include the
locator header fields
c. replace the source address in the IP header with the ALOC
prefix of the locator header
d. replace the destination address in the IP header with the ELOC
prefix of the locator header
e. replace the ALOC prefix in the locator header with the
destination address of the IP header
f. replace the ELOC prefix in the locator header with the source
address of the IP header
g. set the S-field to 1
h. decrease TTL value with one
i. calculate IP-, locator- and transport protocol header
checksums, transport header calculation do not include the
locator header fields
j. forward the packet upon the destination address of the
IP header
4. The swapped hIPv4 packet is now routed inside the remote ALOC
realm upon the new destination address of the IP header to the
final destination.
5. The responder receives the hIPv4 packet
a. the hIPv4 stack must verify the received packet that it uses
the hIPv4 protocol value in the protocol field of the IP
header
Frejborg Expires March 16, 2011 [Page 19]
Internet-Draft Hierarchical IPv4 Framework September 2010
b. verify IP-, locator- and transport protocol header checksums,
transport protocol header verification do not include the
locator header fields
6. The hIPv4 stack of the responder must present to the extended IPv4
socket API the following:
a. present source address as the remote ALOC prefix
b. present destination address as the local IP address
c. verify that the received ALOC prefix equals the local ALOC
prefix
d. present ELOC as the remote IP address
7. The responder's application will respond to the initiator and the
returning packet will take almost the same steps, which are steps
1 to 6, as when the initiator started the session. In step 1 the
responder doesn't need to do a DNS lookup since all information is
provided by the packet.
9. Overlapping Source and Destination ELOC prefixes/ports
Because an ELOC prefix is only significant within the local ALOC
realm there is a slight possibility that a session between two
endpoints residing in separate ALOC realms might use the same source
and destination ELOC prefixes. But the session is still unique
because the two processes communicating over the transport protocol
form a logical session which is uniquely identifiable by the five
tuples involved, i.e. by the combination of <protocol, local IP
address, local port, remote IP address, remote port>.
The session might no longer be unique when two initiators with the
same source ELOC prefix residing in two separate ALOC realms are
accessing a responder located in a third ALOC realm. In this scenario
a possibility exists that the initiators will use the same local port
value. This situation will cause an "identical session situation" for
the application layer. To overcome this scenario the hIPv4 stack must
accept only one unique session with the help of the ALOC information.
If there is an "identical session situation" - i.e. both initiators
uses the same values in the five tuples <protocol, local IP address,
local port, remote IP address, remote port> - the hIPv4 stack shall
allow only the first established session to continue, the following
sessions must be prohibited and the initiator is informed by ICMP
notification about the "identical session situation".
Frejborg Expires March 16, 2011 [Page 20]
Internet-Draft Hierarchical IPv4 Framework September 2010
MPTCP introduces a token, which is locally significant and currently
defined as 32 bit long. The token will provide a sixth tuple for
future stacks to identify and verify the uniqueness of a session -
the probability to have an "identical session situation" is further
reduced. By adding a host identifier scheme to the hIPv4 framework
the "identical session situation" is completely removed.
10. Traceroute considerations
As long as the traceroute is executed inside the local ALOC realm
normal IPv4 traceroute mechanism can be used. As soon as the
traceroute exits the local ALOC realm the locator header shall be
used in the notifications. Therefore extension to ICMP protocol shall
be implemented, the extensions shall be compatible with [RFC4884].
11. Multicast considerations
Since source ELOC prefixes are only installed in the routing table of
the local ALOC realm there is a constraint with Reverse Path
Forwarding (RPF) which is used to ensure loop-free forwarding of
multicast packets. The source address of a multicast group (S,G) is
used against the RFP check. The source address can no longer be used
as a RFP checkpoint outside the local ALOC realm.
To enable RPF globally for a (S,G), the multicast enabled LSR (mLSR)
must at the source's ALOC realm replace the source address with the
local ALOC prefix for inter-ALOC multicast streams. This can be
achieved if the local LSR act also as an anycast Rendezvous Point
with Multicast Source Discovery Protocol (MSDP) and Protocol
Independent Multicast capabilities; with these functionalities the
LSR becomes a multicast enabled LSR (mLSR). The source register at
the mLSR and a source tree is established between the source and the
mLSR. When an inter-ALOC realm receiver subscribes to the multicast
group the mLSR have to swap the hIPv4 header in the following way:
a. verify the received packet that it uses the hIPv4 protocol
value in the protocol field of the IP header
b. verify IP-, locator- and transport protocol header checksums
c. replace the source address in the IP header with the local
ALOC prefix
d. set the S-field to 1
e. decrease TTL value with one
Frejborg Expires March 16, 2011 [Page 21]
Internet-Draft Hierarchical IPv4 Framework September 2010
f. calculate IP-, locator- and transport protocol header
checksums, transport protocol header calculation do not
include the locator header fields
g. forward the packet to the shared multicast tree
In order for the mLSR to function as described above the source must
assemble the multicast hIPv4 packet in the following way:
a. set local IP address (S) from API in the source address
and the ELOC field
b. set multicast address (G) from API in the destination address
field
c. set local ALOC prefix in the ALOC field
d. set the transport protocol value in the protocol field of the
locator header and set the hIPv4 protocol value in the
protocol field of the IP header
e. set the desired parameters in the A-, I-, S-, FI-, VLB-, and
L-fields of the locator header
f. calculate IP-, locator- and transport protocol header
checksums, transport protocol header calculation do not
include the locator header fields. When completed the packet
is transmitted.
The downstream routers from the mLSR towards the receiver will use
the source address (which value is the source's ALOC prefix after the
mLSR) in the IP header for RPF verification. In order for the
receiver to create RTCP receiver reports all information is provided
in the hIPv4 header of the packet.
Because Source Specific Multicast (SSM) and IGMPv3 uses IP addresses
in the payload both protocols needs to be modified to support the
hIPv4 framework.
12. Traffic engineering considerations
When hIPv4 framework is fully implemented ingress load balancing to
an ALOC realm can be influenced by the placement of LSRs at the
realm; a LSR provides a shortest path scheme. Also, if RIR policies
allows, a service provider can have several ALOC assigned, hence
traffic engineering and filtering can be done with the help of ALOC
prefixes. E.g. sensitive traffic can be aggregated under one ALOC
Frejborg Expires March 16, 2011 [Page 22]
Internet-Draft Hierarchical IPv4 Framework September 2010
prefix which is not fully distributed into the DFZ. If needed an ALOC
Traffic Engineering solution between ALOC realms might be developed,
i.e. create explicit paths that can be engineered via specific ALOC
prefixes, e.g. create mechanism similar as described in [Pathlet
Routing]. Further studies are needed; first it should be evaluated if
there is demand for such a solution.
A use case of the ingress load balancing to an ALOC realm can be
described as an "remote ALOC entry routing" solution for a multi-
homed site, i.e. ingress traffic flow to the site is influenced by
how many attachment points to Internet the site uses and where the
attachment points are placed at the local network. In order to apply
"local ALOC exit routing" from the multi-homed site some new network
nodes are needed between the initiator and the border routers. The
new network node(s) shall be able to identify hIPv4 packets, based
upon the protocol field in the IP header, and switch the packets to
explicit paths based upon the ALOC prefix in the locator header.
Together with a multipath transport protocol the subflows can be
routed via specific attachment points, i.e. border routers sitting
between the local network and Internet. Multi-homing becomes multi-
pathing, for details, see appendix B.
The usage of multipath enabled transport protocols opens up the
possibility to develop a new design methodology of backbone networks,
i.e. Valiant Load-Balancing [VLB]. If two single-homed endpoints,
using multipath enabled transport protocols and attached to the
network with only one interface/ELOC-prefix, are communicating over
Internet, both subflows will most likely take the shortest path
throughout Internet. I.e. both subflows are established over the same
links and when there is congestion on a link or a failure of a link
both subflows might simultaneously drop packets - the benefit of
multipath is lost. The "subflows-over-same-links" scenario can be
avoided if the subflows are traffic engineered to traverse Internet
on different paths - but this is difficult to achieve by using
classical traffic engineering, such as IGP tuning or MPLS based
traffic engineering. By adding a mechanism to the locator header the
"subflows-over-same-links" scenario might be avoided. If the LSR
functionality is deployed on a Valiant Load-Balancing enabled
backbone node - hereafter called vLSR - and the backbone nodes are
interconnected via logical full meshed connections, Valiant Load-
Balancing can be applied for the subflows. When a subflow has the
appropriated bits set in the VLB-field of the locator header the
first ingress vLSR shall do VLB switching of the subflow. That is,
the ingress vLSR is allowed to do VLB switching of the subflow's
packets if the VLB bits are set to 01 or 11, the S-bit is set to 0
and the local ALOC prefix of the vLSR matches the ALOC-field's
prefix. If there are no ALOC and ELOC fields in the locator header,
Frejborg Expires March 16, 2011 [Page 23]
Internet-Draft Hierarchical IPv4 Framework September 2010
but the other fields' values are set as described above, the vLSR
should apply VLB switching as well for the subflow - because it is an
inter-ALOC realm subflow belonging to a multipath enabled session.
With this combination of parameters in the locator header the subflow
is VLB switched only at the first ALOC realm and the subflows might
be routed throughout the Internet on different paths. If VLB
switching is applied in every ALOC realm it would most likely add too
much latency for the subflows. The VLB switching at the first ALOC
realm will not separate the subflows on the first and last mile links
(single-homed) - if the subflows on the first and last mile links
needs to be routed on separate links the endpoints should be deployed
in a multi-homed environment. Studies on how Valiant Load-Balancing
is influencing on traffic patterns between interconnected VLB [iVLB]
backbone networks has been carried out. Nevertheless, more studies
are needed around Valiant Load-Balancing scenarios.
13. Large encapsulated packets
Adding the locator header to an IPv4 packet in order to create a
hIPv4 packet will increase the size of it but since the packet is
assembled at the endpoint it will not add complications of current
Path MTU Discovery (PMTUD) mechanism in the network. The intermediate
network between two endpoints will not see any difference in the size
of packet; IPv4 and hIPv4 packet sizes are the same from the network
point of view.
14. Mobility considerations
This section will consider two types of mobility solutions, site
mobility and endpoint mobility.
Site mobility definition:
a site wishes to changes its attachment point to the Internet
without changing its IP address block
Today, classical multi-homing is the most common solution for
enterprises that wishes to achieve site mobility. Multi-homing is one
of the key findings behind the growth of the DFZ RIB - see [IAB
report], sections 2.1 and 3.1.2. The hIPv4 framework can provide a
solution for enterprises to have site mobility without the
requirement of implementing a classical multi-homed solution. This
new multi-homed solution utilizing PI addresses is depended upon the
forthcoming ELOC allocation policy which is discussed in appendix A.
If the regional based ELOC allocation policy is enforced the
enterprise can be concurrently attached to two different Internet
service providers (ISP) without the need to implement AS border
Frejborg Expires March 16, 2011 [Page 24]
Internet-Draft Hierarchical IPv4 Framework September 2010
routing. The ISPs provide their globally unique ALOC prefixes for the
enterprise and the ELOC block of the enterprise is regionally unique,
a PI ELOC block. The enterprise can change the local ALOC prefix at
endpoints - i.e. from the previous ISP's ALOC prefix to the new ISP's
ALOC prefix - without connectivity interruptions inside the local
network. Sessions initiated at the enterprise needs to be routed to
the correct ISP, i.e. the border router (or intermediate routers
between the initiator and border router) of the enterprise needs to
apply local ALOC exit routing. For sessions initiated from the
Internet the DNS record for an endpoint needs to be updated, also the
local ALOC prefix on the endpoint needs be changed to achieve a
symmetric entry/exit path at the site. Since the border router (or
intermediate routers) is enforcing local ALOC exit routing the
responder can apply basic session load balancing over the two ISPs
based upon from which ISP the session has been initiated, i.e. if the
responder have two valid DNS records with two different ALOC
prefixes. Conclusion is that a single-homed enterprise can achieve
smooth transition from one ISP to another by only changing the ALOC
prefix on the endpoints and at DNS records - the local ELOC scheme
remains intact. Also a single-homed enterprise can become multi-homed
without implementing AS border routing or to have an own ALOC prefix
assigned. If a better session load balancing scheme is required the
application should be migrated to a multipath enabled transport
protocol such as [SCTP] or [MPTCP]. Multi-homing is discussed in
detail in appendix B.
Endpoint mobility definition:
an endpoint moves relatively rapidly between different networks,
changing its IP layer network attachment point
Mobile IP [MIP] is used today for endpoints in order to provide
mobility. Mobile IP is an overlay protocol; it is also an application
that uses IP addresses in its payload. It is obvious that hIPv4
extensions need to be added to the MIP framework. Another approach is
to investigate what [MPTCP] can offer to solve endpoint mobility
scenarios. MPTCP introduces a token, which is locally significant and
currently defined as 32 bit long. The token will provide a sixth
tuple to identify and verify the uniqueness of a session. This sixth
tuple - the token - is not depended upon the underlying layer, i.e.
the IP layer. The session is identified with the help of the token
and thus the application is not aware when the IP parameters are
changed, e.g. during a roaming situation - but it is required that
the application is not making use of IP addresses. Security issues
arise; the token can be capture during the session by e.g. a man-in-
the-middle attack - these attacks can be mitigated by applying
[tcpcrypt]. If the application requires full protection against man-
Frejborg Expires March 16, 2011 [Page 25]
Internet-Draft Hierarchical IPv4 Framework September 2010
in-the-middle attacks the user should apply Transport Layer Security
[TLS] Protocol for the session.
To summarize, the most common endpoint mobility use case today is,
that the responder resides in the fixed network and the initiator is
mobile - thus MPTCP will provide roaming capabilities for the mobile
endpoint - if both endpoints are making use of the MPTCP extension.
However, in some use cases the fixed endpoint needs to initialize a
session to a mobile responder. Thus MIP should incorporate the hIPv4
extension - MIP is providing a rendezvous service for the mobile
endpoints. Also many applications are providing rendezvous services
for their users, e.g. SIP, peer-to-peer, Instant Messaging services
etc. A generic rendezvous service solution can be provided by a host
identifier scheme, e.g. [HIP], [ILNP] or [NBS]. If desired the end
user (actually the application) can make use of one of these
rendezvous service schemes, i.e. MPTCP, extensions to MIP, some
application specific rendezvous service or a generic rendezvous
service - or some kind of combination of them. The routing
architecture is only providing location information for the
endpoints, i.e. the identifier mechanisms (session and host
identifiers) are decoupled from the routing architecture.
15. Affected Applications and Implications
There are several applications that are inserting IP address
information in the payload of a packet. Some applications use the IP
address information to create new sessions or for identification
purposes. This section is trying to list the applications that need
to be enhanced; however, this is by no means a comprehensive list.
The applications can be divided in four main categories:
o Applications based on raw sockets, a raw socket is receiving
packets containing the complete header in comparison to the other
sockets that only receives the payload.
o Applications needed to enable the hIPv4 framework, i.e. DNS and
DHCP databases which must be extended to support ALOC prefixes.
o Applications that insert IP addresses to the payload and uses the
IP address for setting up new sessions or for some kind of
identification. The application belonging to this category can not
set up sessions to other ALOC domains until extensions have been
incorporated. Within the local ALOC domain there are no
restrictions since the current IPv4 scheme is still valid. The
following applications have been identified:
Frejborg Expires March 16, 2011 [Page 26]
Internet-Draft Hierarchical IPv4 Framework September 2010
o SIP; IP addresses are inserted in the SDP, Contact and Via
header
o Mobile IP; the mobile node uses several IP addresses during
the registration process
o IPsec AH; designed to detect alterations at the IP packet
header
o RSVP; RSVP messages are sent hop-by-hop between RSVP-capable
routers to construct an explicit path
o ICMP; notifications needs to be able to incorporate ALOC
information and assemble the hIPv4 header in order to be
routed back to the source
o Source Specific Multicast; the receiver must specify the
source address
o IGMPv3; a source-list is included in the IGMP reports
o Applications related to security, such as firewalls, must be
enhanced to support ALOC prefixes
o Applications that will function with FQDN but many uses an IP
addresses instead, such as ping, traceroute, telnet and so on. The
CLI syntax needs to be upgraded to support ALOC and ELOC
information via the extended socket API.
16. The Future Role of the LSR
The LSR was added to the framework in order to provide a smooth
transition from the current IPv4 framework to the hIPv4 framework,
i.e. a major forklift of the current forwarding plane is avoided by
the introduction of the LSR element. In the future, the LSR can be
left as such in the network, if preferred, or the LSR functionality
can be expanded towards the edge when routers are upgraded due to
their natural lifecycle process. Once an upgrade of a router is
required because of e.g. increased demand for bandwidth, the modified
forwarding plane might support concurrently IPv4 and hIPv4 forwarding
- and the LSR functionality can be pushed towards the edge and in the
future removed. This is accomplished by adding extension to the
current routing protocols, both IGP and BGP. When a LSR receives a
hIPv4 packet where the destination IPv4 address matches the local
ALOC prefix the LSR shall - contrary to the tasks defined in section
8, step 4 - lookup the ELOC field in the locator header and compare
this prefix against the FIB. If the next-hop entry is LSR capable the
Frejborg Expires March 16, 2011 [Page 27]
Internet-Draft Hierarchical IPv4 Framework September 2010
packet shall be forwarded upon the ELOC prefix. If the next-hop is a
classical IPv4 router the LSR must apply the tasks defined in section
8, step 4 and once completed forward the packet upon the new IPv4
destination address.
Once the routers from the first ingress LSR to the final destination
endpoint is upgraded to support hIPv4 forwarding there exist no
longer a need to implement LSR functionality in the network, the
packet is forwarded as such to the endpoint's extended stack. The
hIPv4 stack must check that the ELOC prefix matches its local IPv4
address, because the destination IPv4 address matched the local ALOC
prefix. Then the hIPv4 stack of the destination must present to the
extended IPv4 socket API the following:
a. present source address as the remote IP address
b. present the destination address as the local ALOC prefix
c. present the ALOC prefix as the remote ALOC prefix
d. present the ELOC prefix as the local IP address
When all endpoints (that need to establish sessions outside the local
ALOC realm) and infrastructure nodes are hIPv4 capable there is no
need to apply LSR functionality for either unicast or multicast
sessions - forwarding decisions can be made upon information in the
IP and locator header. In the local ALOC realm packets are routed to
their upstream anycast or unicast ALOC border router upon the ALOC
prefix in the locator header - i.e. a local ALOC exit routing scheme
is applied upon the local ALOC FIB. Since the anycast or unicast ALOC
node is the owner of the local ALOC prefix it will forward the
packets upon the destination IP prefix - local ALOC exit routing
halts at the border router and the FI-bits are changed to 01. At the
intermediate ALOC realms the forwarding is based upon the packet's
destination address in the IP header and the global DFZ FIB is used.
Finally the packet arrives to the remote ALOC realm's border router -
which is the owner of the remote ALOC prefix and matches the
destination address in the IP header, thus the FI-bits are changed to
11. Hereafter forwarding of the packet is applied upon the ELOC
prefix in the locator header, i.e. the remote ELOC FIB is used. Note
that IP and transport protocol headers will remain intact, only FI
and Locator Header Checksum values in the locator header will
alternate. The new forwarding plane will enable a 32 x 32 bit locator
space, for further details see appendix A.
Also multicast LSR (mLSR) functionality can be removed when the
forwarding plane is upgraded. The source's ALOC border router set the
FI-bits to 10 and RFP check is hereafter applied on the ALOC field in
Frejborg Expires March 16, 2011 [Page 28]
Internet-Draft Hierarchical IPv4 Framework September 2010
the locator header - also here IP and transport protocol headers will
not alternate.
For intra-ALOC realm sessions classical IPv4 forwarding plane is kept
in place.
17. Transition considerations
The hIPv4 framework is not introducing any new protocols that would
be mandatory to carry out the transition from IPv4 to hIPv4; instead
extensions are added to existing protocols - the hIPv4 framework
requires extensions to the current IPv4 stack, infrastructure systems
and to some applications that use IP addresses in the payload but the
current forwarding plane in Internet remains intact (short term)
apart from that a new forwarding element (the LSR) is required to
create an ALOC realm. Extensions to the IPv4 stack, infrastructure
systems and applications that uses IP addresses in the payload and
routers can be deployed in parallel with the current IPv4 framework.
Even genuine hIPv4 sessions can be established between endpoints
though the current single dimensional Internet structure is still
present. When will the single dimensional routing architecture then
be upgraded to a hierarchical architecture? The author thinks there
are two possible tipping points:
o When the RIB of DFZ is getting close to the capabilities of
current forwarding plane - who will pay for the upgrade? Or will
the service provider only accept ALOC prefixes from other service
providers and avoid capital expenditures?
o When the depletion of IPv4 addresses is causing enough problems
for service providers and enterprises.
The biggest risk why hIPv4 framework will not succeed is the short
timeframe before the expected depletion of the IPv4 address space
occurs. Also, will enterprise give up their global allocation of the
current IPv4 address block they have gained? Another risk is, will
the enterprises and residences carry out an upgrade of their
endpoints and security nodes? Transitions arguments and methods are
discussed in appendix C and D.
18. Security Considerations
Hijacking of a single ELOC prefix by longest match from another ALOC
realm is no longer possible since the prefixes are separated by a
locator, the ALOC. To apply a hijack of a certain ELOC prefix the
whole ALOC realm must be routed via a bogus ALOC realm. Studies
should be carried out with the Secure Inter-Domain Routing (SIDR)
workgroup if the ALOC prefixes can be protected from hijacking.
Frejborg Expires March 16, 2011 [Page 29]
Internet-Draft Hierarchical IPv4 Framework September 2010
19. IANA Considerations
There are no IANA considerations for this document.
20. Conclusion
This document provides a high level overview of the hierarchical IPv4
framework which could be build in parallel with the current single
dimensional Internet by implementing extensions at several
architectures. Implementation of the hIPv4 framework will not require
a major service window break in the Internet, neither at the private
networks of enterprises. Basically, the hIPv4 framework is an
evolution of the current IPv4 framework. For sessions inside an ALOC
realm the classical IPv4 framework can be used in the future and for
inter-ALOC realm sessions the hIPv4 framework is needed. Though there
is a long journey ahead and many things that need to be sorted out
the hierarchical IPv4 framework looks promising. The transition can
be attractive for the enterprises since the hIPv4 framework doesn't
create a catch-22 situation, it introduces functionalities (related
to site and endpoint mobility, better long distance performance, new
tools to design and implement split-DNS architectures) that could
better serve their business models, introduce less expensive multi-
homing solutions, it slows down the expected growth of Internet's
carbon footprint and it is inline with the Corporate Social
Responsibility programs that many enterprises have implemented. The
framework should also be interesting for the service providers, when
the transition phase is completed the growth of DFZ will be
controlled by the service providers and only the service providers -
multi-homed enterprises do not influence on the RIB size of the DFZ
anymore. After the transition the RIB size of the DFZ will be
reduced, which should have a decreasing effect on the expected cost
structure of future DFZ routers, both operating and capital
expenditure. A long term evolution will provide a 32 x 32 bit locator
space, the ALOC prefixes are allocated only by service providers -
ELOC prefixes are only significant at a local ALOC realm, i.e. an
enterprise can use a 32 bit locator space for its private network
(the ALOC prefix is rented from the attached ISP) and an ISP can use
a 32 bit ELOC space (for each ALOC prefix) to provide Internet
connectivity services for its directly attached customers
(residential and enterprise). An identifier placeholder has been
added to the locator header; a host identifier scheme can be added
but is not mandatory to enable it - a multipath transport protocol
leverages a session identifier mechanism that provide mobility
services and in some use cases an identifier mechanism is not needed
at all.
Frejborg Expires March 16, 2011 [Page 30]
Internet-Draft Hierarchical IPv4 Framework September 2010
21. References
21.1. References
[IAB report]
Meyer, D., Zhang, L., Fall, K., "Report from the IAB
Workshop on Routing and Addressing", RFC 4984, September
2007
[MPLS] Rosen, E., Tappan, D., Fedorkow, G., Rekther, Y.,
Farinacci, D., Li, T., Conta, A., "MPLS Label Stack
Encoding", RFC 3032, January 2001
[RFC4884] Bonica, R., Gan, D., Tappan, D., Pignataro, C. "Extended
ICMP to support Multi-Part Messages", RFC 4884, April 2007
[RFC1918] Rekhter, Y., Moskowitz, B., Karrenberg, D., de Groot, G.J.,
Lear, E. "Address Allocation for Private Internets", RFC
1918, February 1996
[HIP] Moskowitz, R., Nikander, P. "Host Identity Protocol (HIP)
Architecture", RFC 4423, May 2006
[SCTP] Stewart, R. "Stream Control Transmission Protocol", RFC
4960, September 2007
[MIP] Perkins, C. "IP Mobility Support for IPv4", RFC 3344,
August 2002
[EIP] Wang, Z. "The Extended Internet Protocol", RFC 1385,
November 1992
[TLS] Dierks, T., Rescorla, E., "The Transport Layer Security
(TLS) Protocol Version 1.2", RFC 5246, August 2008
[RFC1812] Baker, F., "Requirements for IP Version 4 Routers", RFC
1812, June 1995
[DNSSEC] Arends, R., Austein, R., Larson, M., Massey, D., Rose, S.,
"DNS Security Introduction and Requirements", RFC 4033,
March 2005
21.2. Informative References
[RBridge] Perlman, R., "RBridges, Transparent Routing", Infocomm,
2004
Frejborg Expires March 16, 2011 [Page 31]
Internet-Draft Hierarchical IPv4 Framework September 2010
[Dagstuhl]
Arkko, J., Braun, M.B., Brim, S., Eggert, L., Vogt, C.,
Zhang, L., "Perspectives Workshop: Naming and Addressing in
a Future Internet", Dagstuhl, 2009
[Nimrod] Chiappa N., "A New IP Routing and Addressing Architecture",
1991
[MPTCP] Ford, A., Raiciu, C., Handley, M., Barre, S., "TCP
Extensions for Multipath Operation with Multiple
Addresses", IETF TSV Area, July 2009
[VLB] Zhang-Shen, R., McKeown, N., "Designing a Predictable
Internet Backbone with Valiant Load-Balancing", Stanford
University
[iVLB] Babaioff, M., Chuang, J., "On the Optimality and
Interconnection of Valiant Load-Balancing Networks",
University of California at Berkeley
[RRG] RRG, "IRTF Routing Research Group Home Page",
http://tools.ietf.org/group/irtf/trac/wiki/RoutingResearchG
roup
[BFD] Bidirectional Forwarding Detection Workgroup,
http://www.ietf.org/dyn/wg/charter/bfd-charter.html
[CES] Jen, D., Meisel, M., Yan, H. Massey, D., Wang, L., Zhang,
B., Zhang, L., "Towards A New Internet Routing
Architecture: Arguments for Separating Edges from Transit
Core", Sigcomm
[ILNP] Atkinson, R. "ILNP Concept of Operations", ILNP Intro
[NBS] Vogt, C. "Simplifying Internet Applications Development
With A Name-Based Sockets Interface", christianvogt
[Pathlet Routing]
Godfrey, P.G., Shenker, S., Stoica, I., "Pathlet Routing",
Sigcomm
[tcpcrypt]
Bittau, A., Hamburg, M., Handley, M., Mazi`eres, D., Boneh,
D., "The case for ubiquitous transport-level encryption",
Usenix
Frejborg Expires March 16, 2011 [Page 32]
Internet-Draft Hierarchical IPv4 Framework September 2010
[LISP] Farinacci, D., Fuller, V., Meyer, D., Lewis, D.,
"Locator/ID Separation Protocol", LISP WG
[RRG recommendation]
Li, T., "Recommendation for a Routing Architecture", IRTF
Routing RG
22. Acknowledgements
The active participants at the Routing Research Group [RRG] mailing-
list are acknowledged. The participants have provided ideas,
proposals and discussions that have influenced on the architecture of
the hIPv4 framework.
Frejborg Expires March 16, 2011 [Page 33]
Internet-Draft Hierarchical IPv4 Framework September 2010
Appendix A. Short term & future IPv4 address allocation policy
In this section we will discuss and study how the hIPv4 framework
could influence on the IPv4 address allocation policies to ensure
that the new framework will enable some re-usage of IPv4 address
blocks. It is the Regional Internet Registries (RIRs) that shall
define the final policies.
When the hIPv4 framework is fully implemented (assume current
forwarding plane is still in place) every ALOC realm could have a
full IPv4 address space - except the GLB - to allocate ELOC blocks
from. There are some implications though. In order for an enterprise
to achieve site mobility, i.e. to change service provider without
changing its ELOC scheme, the enterprise should implement an
Autonomous System (AS) solution with ALOC prefix at the attachment
point to the service provider. Larger enterprises do have the
resources to implement AS border routing; most of the large
enterprises have already implemented multi-homing solutions. The
small and midsize enterprises (SME) may not have the resources to
implement AS border routing, or the implementation introduces
unnecessary costs for the SME. Also if every enterprise needs to have
an ALOC prefix it will have an impact on the RIB at the DFZ, the RIB
will be populated with a huge amount of ALOC prefixes.
It is clear that a compromise is needed. A SME is usually single-
homed and the SME should be able to reserve a PI ELOC block from the
RIR without the need to be forced to create an ALOC realm, i.e.
implement a LSR solution and AS border routing. The PI ELOC block is
no longer globally unique, the SME can only reserve the PI ELOC block
for the region where it is active or has it attachment point to
Internet. The attachment point rarely changes to another country;
therefore it is sufficient that the PI ELOC block is regionally
unique. When the enterprise is replacing its Internet service
provider the enterprise do not have to change its ELOC scheme - only
the local ALOC prefix at the endpoints are changed. The internal
traffic at an enterprise does not use the ALOC prefix, the internal
routing is applied by the IP header and thus the internal routing and
addressing architectures are preserved.
Mergers and acquisitions of enterprise can cause ELOC conflicts,
because the PI ELOC block is hereafter only regionally unique. If an
enterprise in region A overtakes an enterprise in region B there is a
slight chance that both enterprises can have overlapping ELOC spaces.
If overlapping of ELOC spaces occurs the private unicast ALOC
solution can be implemented to separate them - if all affected
endpoints support the hIPv4 framework.
Frejborg Expires March 16, 2011 [Page 34]
Internet-Draft Hierarchical IPv4 Framework September 2010
Finally, residential users will receive only PA locators. When a
residential user changes a service provider the residential user has
to replace the locators. A PA ELOC block is no longer globally
unique, every Internet service provider can use the PA ELOC blocks at
their ALOC realms - the PA locators becomes kind of private locators
for the service providers.
If the forwarding plane and all hosts, that establishes inter-ALOC
realm sessions, are upgraded to support the hIPv4 framework (it will
most likely take decades) several interesting possibilities occurs:
o the regional allocation policy for PI ELOC spaces can be removed,
the enterprise can make use of the whole IPv4 address space that
is globally unique today - the ELOC space is hereafter only
significant at a local ALOC realm.
o in case of mergers or acquisitions of enterprises the private
unicast ALOC solution can be used to separate overlapping ELOC
spaces.
o the GLB space can be expanded to make use of all 32 bits (since
enterprises have no interest or benefits to apply for globally
unique locators) for anycast and unicast ALOC allocations, only
ISP are allowed to apply for GLB prefixes.
o the anycast ALOC solution can be replaced with the global unicast
ALOC solution since the ISP and enterprise no longer need to share
ELOC routing information between each other. Also there is enough
space in the GLB to allocate global unicast ALOC for every
enterprise.
o residential users will still use global anycast ALOC solutions and
if they change service provider their locators needs to be
replaced.
Outcome is that a 32 x 32 bit locator space is achieved, when an
enterprise is replacing an ISP with another ISP only the ALOC
information is replaced at endpoints and infrastructure nodes -
renumbering of ALOC prefixes can be automated by e.g. DHCP and
extensions to IGP.
Frejborg Expires March 16, 2011 [Page 35]
Internet-Draft Hierarchical IPv4 Framework September 2010
Appendix B. Multi-homing becomes multi-pathing
When the transition of the hIPv4 framework is fully completed (except
forwarding plane) the RIB of an ISP, that has created an ALOC realm,
will have the following entries:
o the PA ELOC blocks of directly attached customers (e.g.
residential and enterprises)
o the PI ELOC blocks of directly attached customers
o the globally unique ALOC prefixes, received from other service
providers
The ISP will not carry in its RIB any PA or PI ELOC blocks from other
service providers. In order to do routing and forwarding of packets
between ISPs only ALOC information of other ISPs is needed.
Then the question is, how to keep the growth of ALOC reasonable - if
the enterprise is using PI addresses, having an AS number and
implementing BGP, why not apply for an ALOC prefix?
The classical multi-homing is causing the biggest impact on growth of
the size of the RIB in the DFZ - replacing a /20 IPv4 prefix with a
/32 ALOC prefix will not reduce the size of the RIB in the DFZ.
Most likely the only way to prevent this from happening is to have a
yearly cost for the allocation of an ALOC prefix - except if you are
a service provider that are providing access and/or transit traffic
for your customers. And it is granted to have cost for allocating an
ALOC prefix for the non-service providers, because when an enterprise
is using an ALOC prefix the enterprise is reserving a FIB entry
throughout the DFZ - and the ALOC FIB entry needs to have power,
space, hardware and cooling on all the routers in the DFZ.
By implementing this kind of ALOC allocating policy it will reduce
the RIB size in the DFZ quite well, multi-homing will no longer
increase the RIB size of the DFZ. But this policy will have some
impact on the resilience behavior, by compressing routing information
we will loose visibility in the network. In today's multi-homing
solutions the network always know where the remote endpoint resides,
in case of a link or network failure a backup path is calculated and
an alternative path is found - all routers in the DFZ are aware of
the change in the topology. This functionality has off-loaded the
workload of the endpoints; they only need to find the closest ingress
router and the network will deliver the packets to the egress router,
regardless of what failures (almost) happen in the network. And with
Frejborg Expires March 16, 2011 [Page 36]
Internet-Draft Hierarchical IPv4 Framework September 2010
the growth of multi-homed prefixes the routers in the DFZ have been
forced to carry a greater workload, perhaps close to their limits -
the workload between the network and endpoints is not in balance.
Conclusion is that the endpoints should take more responsibilities
for their sessions and that way off-loads the workload in the
network. How, lets walkthrough an example:
A remote enterprise has been given an ELOC block 192.168.1.0/24 that
is either via static routing or BGP announced to the upstream service
providers. The upstream service providers are providing the ALOC
information for the enterprise, i.e. 10.1.1.1 and 10.2.2.2. A remote
endpoint has been installed, it has been given ELOC 192.168.1.1 - the
ELOC is a locator defining where the remote endpoint is attached to
the remote network. The remote endpoint has been assigned ALOCs
10.1.1.1 and 10.2.2.2 - an ALOC is a locator defining the attachment
point of the remote network to Internet.
The initiator (local endpoint), which has ELOC 172.16.1.1 and ALOC
prefixes 10.3.3.3 & 10.4.4.4, has established a session by using
source ALOC 10.3.3.3 to the responder (remote endpoint) at ELOC
192.168.1.1 and ALOC 10.1.1.1. I.e. both networks 192.168.10/24 and
172.16.1.0/24 are multi-homed. ALOC are not available in current IP
stack's API but both ELOC are seen as the local and remote IP
addresses in the API, so the application will communicate between IP
addresses 172.16.1.1 and 192.168.1.1 - if ALOC prefixes are included
the session is established between 10.3.3.3:172.16.1.1 and
10.1.1.1:192.168.1.1.
Next a network failure occurs, the link between the responder border
router (BR-R1) and service provider that owns ALOC 10.1.1.1 goes
down. The border router of the initiator (BR-I3) will not be aware of
the situation, because only ALOC information is exchanged between
service providers and ELOC information is compressed to stay within
ALOC realms. But BR-R1 will notice the link failure; BR-R1 could
rewrite the ALOC field in the locator header for this session, i.e.
from 10.1.1.1 to 10.2.2.2 and send the packets to the second service
provider via BR-R2 - the session between the initiator
10.3.3.3:172.16.1.1 and the responder 10.2.2.2:192.168.1.1 remains
intact because the five legacy tuples at the IP stack API do not
change - only the ALOC prefix of the responder has changed and this
information is not shown to the application. An assumption here is
that the hIPv4 stack does accept changes of ALOC prefixes on the fly
(more about this later).
If the network link between the BR-I3 and ISP providing ALOC 10.3.3.3
fails, BR-I3 could rewrite the ALOC prefix in the locator header and
route the packets via BR-I4 - and the session stays up. If there is a
Frejborg Expires March 16, 2011 [Page 37]
Internet-Draft Hierarchical IPv4 Framework September 2010
failure somewhere in the network the border routers might receive an
ICMP destination unreachable message (if not blocked by security
functionality) and thus try to switch to session over to the other
ISP by replacing the ALOC prefixes in the hIPv4 header. Or the
endpoints might try themselves to switch to the other ALOCs after a
certain time-out in the session. In all session transition cases the
five legacy API tuples remains intact.
If border routers or one of the endpoint changes the ALOC prefix
without a negotiation with the remote endpoint security issues
arises. Can the endpoint(s) trust the remote endpoint when ALOC
prefix(es) are changed on the fly - is it still the same remote
endpoint or has the session been hijacked by a bogus endpoint? The
obvious answer is that an identification mechanism is needed to
ensure that after a change in the path or a change of the attachment
point of the endpoint the endpoints are still the same. An identifier
needs to be exchanged during the transition of the session. Two types
of identifiers have been discussed on the [RRG] mailing-list, session
and host identifiers. Both types of identifiers can be used to
protect the session from being hijacked. A session identifier will
provide a low level security mechanism, offering some protection
against hijacking of the session and also provide mobility. [SCTP]
uses the verification tag to identify the association; [MPTCP]
incorporates a token functionality for the same purpose - both can be
considered to fulfill the characteristics of a session identifier.
[tcpcrypt] can be used to further mitigate session hijacking. If the
application requires full protection against man-in-the-middle
attacks the user should apply Transport Layer Security [TLS] Protocol
for the session. Both transport protocols are also multipath capable.
Implementing multipath capable transport protocols in a multi-homed
environment will provide new capabilities such as:
o concurrent and separate exit/entry paths via different attachment
points at multi-homed sites
o true dynamic load-balancing, the endpoints do not participate in
any routing protocols or do not update rendezvous solutions due to
network link or node failures
o only a single NIC on the endpoints is required
o in case of a border router or ISP failure, the multipath transport
protocol will provide resilience
By adding more intelligence at the endpoints, i.e. multipath enabled
transport protocols, the workload of the network is off-loaded and
can take less responsibility for providing visibility of destination
Frejborg Expires March 16, 2011 [Page 38]
Internet-Draft Hierarchical IPv4 Framework September 2010
prefixes in Internet - i.e. prefix compression in the DFZ can be
applied and only the attachment points of a local network needs to be
announced in the DFZ. And the IP address space no longer needs to be
globally unique; it is sufficient that only a part is globally
unique, the rest is only regionally unique (long term, locally
unique) as discussed in appendix A. Outcome is that the current
multi-homing solution can migrate towards a multi-pathing environment
that will have the following characteristics:
o AS number is not mandatory for enterprises
o BGP protocol is not mandatory at the enterprise's border routers,
static routing with Bidirectional Failure Detection [BFD] is an
option
o allocation of global ALOC prefix(es) for the enterprise should not
be allowed, instead upstream ISPs are providing the global ALOC
prefixes for the enterprise
o MPTCP provides dynamic load-balancing without using routing
protocols, several paths can be simultaneously used and thus
resilience is achieved
o provide low growth of RIB entries at the DFZ
o when static routing is used between the enterprise and ISP:
o the RIB size at the enterprise's border routers are not
depended upon the size of the RIB in DFZ nor adjacent ISPs
o the enterprise's border router can not cause BGP churn in the
DFZ or in the adjacent ISPs' RIB
o when dynamic routing is used between the enterprise and ISP:
o the RIB size at the enterprise's border routers are depended
upon the size of the RIB in DFZ and adjacent ISPs
o the enterprise's border router can cause BGP churn for the
adjacent ISPs but not in the DFZ
o the cost of border router should be less expensive that in today's
multi-homing solution
Frejborg Expires March 16, 2011 [Page 39]
Internet-Draft Hierarchical IPv4 Framework September 2010
Appendix C. Transition Arguments
The media has announced several times the meltdown of Internet and
the depletion of IPv4 addresses - but the potential chaos has been
postponed several times and the general public has lost their
interest in these announcements. Perhaps other approaches could be
worthwhile to study, instead try to find other valuable arguments
that the general public could be interested in, such as:
o Not all endpoints needs to be upgraded, only those endpoints that
are directly attached to the Internet. These kinds of endpoints
are portable laptops, smart mobile phones, proxies, and
DMZ/frontend endpoints. But the most critical endpoints, the
backend endpoints where enterprises keep their most critical
business applications do not need to be upgraded; the backend
endpoints should not be reached at all from Internet - only from
the private network - and this functionality can be achieved with
the hIPv4 framework, since it is backwards compatible with the
current IPv4 stack.
o Mobility, it is estimated that the demand for applications that
performs well over the wireless access network will increase.
Introduction of MPTCP and host identifier schemes opens up a new
possibility to create new solutions and applications that are
optimized for mobility. The hIPv4 framework requires an upgrade of
the endpoints' stack; if possible the hIPv4 stack should also
contain MPTCP and host identifier scheme features. Applications
designed for mobility could bring competitive benefits for the
enterprises.
o The intermediate routers in the network do not need immediately to
be upgraded; the current forwarding plane can still be used. The
benefit is that the current network equipment can be preserved at
the service providers, enterprises and residences (except
middleboxes). That means that the carbon footprint is a lot lower
compared to other solutions. Many enterprises do have green
programs and many residential users are concerned with the global
warming issue.
Frejborg Expires March 16, 2011 [Page 40]
Internet-Draft Hierarchical IPv4 Framework September 2010
o The migration from IPv4 to IPv6 (current defined architecture)
will increase the RIB and FIB throughout DFZ, will it require a
new upgrade of the forwarding plane as discussed in the IAB report
is unclear. Most likely an upgrade is needed, the outcome of
deploying IPv4 and IPv6 concurrently is that the routers need to
have larger memories for the RIB and FIB - every globally unique
prefix is installed in the routers that are participating in the
DFZ. Since the enterprise is reserving one or several RIB/FIB
entries on every router in the DFZ it means that the enterprise is
increasing the power consumption of Internet, thus increasing the
carbon footprint. And many enterprises are committed to green
programs - if hIPv4 gets deployed, the power consumption of
Internet will not grow as much as compared in an IPv4 to IPv6
transition scenario.
o Another issue, if the migration from IPv4 to IPv6 (current defined
architecture) occurs, is that the routers in the DFZ most likely
need to be upgraded to more expensive routers - as discussed in
the IAB report. In the wealthy part of the world, where a large
penetration of Internet users is already present, the service
provider can pass along more easily the costs of the upgrade to
their subscribers - with a "wealthy/high penetration" ratio the
cost will not grow that much that the subscribers would abandon
Internet. But in the less wealthy part of the world, where there
is usually a lower penetration of subscribers, the cost of the
upgrade cannot that easily be covered - a "less wealthy/low
penetration" ratio could have a dramatic increase on the cost that
needs to be passed along to the subscribers. And thus fewer
subscribers could afford to get connected to the Internet. For the
global enterprises and the enterprises in the less wealthy part of
the world, this scenario could mean less potential customers and
there could be situations when the nomads of the enterprises can't
get connected to Internet. This is also not fair; every human
being should have a fair chance to be able to enjoy the Internet
experience - and the wealthy part of the world should take this
right into consideration. Many enterprises are committed to
Corporate Social Responsibility programs.
Not only technical and economical arguments can be found, also other
arguments that the general public is interested in and concerned
about can be found. Such arguments as that the Internet becomes
greener and more affordable for everyone than compared to current
forecast of the evolution of Internet. These non-technical values
need to be communicated to the general public, you could ask them -
do you care about the Planet, People and Internet? If you do, please
upgrade the stack on your Internet enabled device.
Frejborg Expires March 16, 2011 [Page 41]
Internet-Draft Hierarchical IPv4 Framework September 2010
Appendix D. Integration with CES architectures
Because the hIPv4 framework requires changes to the endpoints' stack
it will take some time before the migration of the current IPv4
framework to a hIPv4 enabled routing architecture is fully completed.
If a hIPv4 proxy solution could be used in front of classical IPv4
endpoints, the threshold for early adopters to start to migrate
towards the hIPv4 framework would be less questionable and the
migration phase most likely would also be much shorter. Thus it
should be investigated if the hIPv4 framework can be integrated with
Core-Edge Separation [CES] architectures. In a CES architecture the
endpoints do not need to be modified - the main design goal of a CES
solution is to minimize the PI-address entries in the DFZ and to
preserve the current stack at the endpoints. But a CES solution
requires a new mapping system and will also introduce a caching
mechanism in the map-and-encapsulate network nodes. Much debate about
scalability of a mapping system and the caching mechanism has been
carried out at the [RRG] list - today it is unclear how well both
solutions will scale - research work of both topics are still in
progress.
Since the CES architectures are dividing the address spaces in two
new categories - one that is installed in the RIB of the DFZ and the
other that is installed in the local networks - there are to some
degree similarities between CES architectures and the hIPv4
framework. Actually, the invention of the LSR functionality was
inspired by [LISP].
In order to describe how these two architectures might be integrated
some terminology definitions are needed:
CES-node:
A network node installed in front of a local network that must have
the following characteristics
o map-and-encapsulate ingress functionality
o map-and-encapsulate egress functionality
o incorporate the hIPv4 stack
o routing functionality, [RFC1812]
o be able to apply policy based routing on the ALOC field in the
locator header
Frejborg Expires March 16, 2011 [Page 42]
Internet-Draft Hierarchical IPv4 Framework September 2010
Note that the CES-node do not include the [MPTCP] extension, it
would most likely put too much burden on the CES-node to signal and
maintain MPTCP subflows for the cached hIPv4 entries.
Consumer site:
A site that is not publishing any services towards Internet, i.e.
there are no entries in DNS for this site. This site is used by
local endpoints to establish outbound connectivity, i.e. endpoints
are initiating sessions from the site towards content sites.
Usually this site is found at small enterprises and residencies.
PA-addresses are usually assigned to this type of site.
Content site:
A site that is publishing services towards Internet, usually these
services do have DNS entries. This site is used by local endpoints
to establish both inbound and outbound connectivity. The large
enterprises are using PI-addresses and the midsize/small
enterprises are using either PI- or PA-address space.
The CES architectures are aiming to reduce the PI-address entries in
the DFZ, thus map-and-encapsulate egress functionality shall be
installed in front of the content sites (map-and-encapsulate ingress
functionality is required at the Internet Service Providers, ISP, but
for the hIPv4-CES integration study the map-and-encapsulate ingress
functionality at ISP are not interesting - but LSR functionality and
provider map-and-encapsulate ingress functionality might reside in
the same node). It is likely that the node containing map-and-
encapsulate egress functionality will also contain map-and-
encapsulate ingress functionality; it is most likely a router so the
node just needs to support the hIPv4 stack and be able to apply
policy-based routing upon the ALOC field of the locator header to
become a CES-node. It is possible that the Large Content Providers
(LCP) are not willing to install map-and-encapsulate functionality in
front of their sites - if the caching mechanism is not fully reliable
or if the mapping lookup delay do have an impact on their customers'
end user experience then most likely the LCP will not adopt the CES
architecture.
In order to convince a LCP to adopt the CES architecture, it should
provide a mechanism to mitigate the caching and mapping lookup delay
risks. One method is to push the CES architectures to the edge - the
closer to the edge you add new functionality the better it will
scale, that is, if the endpoint stack is upgraded the caching
mechanism is maintained by the endpoint itself. The mapping mechanism
can be removed if the CES architecture's addressing scheme is
Frejborg Expires March 16, 2011 [Page 43]
Internet-Draft Hierarchical IPv4 Framework September 2010
replaced with the addressing scheme of hIPv4 when the CES solution is
integrated at the endpoints. With this approach the LCP might install
a CES-node in front of their sites; also some endpoints within the
content site might be upgraded with the hIPv4 stack.
If the LCP faces issues with the caching or mapping mechanisms the
provider can ask its customers to upgrade their endpoints' stack to
ensure a proper service level. At the same time the LCP promotes the
migration from the current routing architecture to a new routing
architecture, not for the sake of the routing architecture but
instead to ensure a proper service level - you can say that a
business model will promote the migration of a new routing
architecture.
The hIPv4 framework proposes that the IPv4 addresses (ELOC) should no
longer be globally unique; once the transition is completed a more
regional allocation can be deployed. But this is only possible once
all endpoints (that are establishing sessions to other ALOC realms)
have migrated to support the hIPv4 framework. Here the CES
architecture can speed up the re-usage of IPv4 addresses, i.e. once
an IPv4 address block has become an ELOC block it can be re-used
within the other RIR regions - also without the requirement that all
endpoints in Internet must first be upgraded. As said earlier the CES
architecture is aiming to remove PI-addresses from the DFZ, thus the
content sites are more or less the primary target for the roll-out of
a CES solution. At large content sites a CES-node most likely will be
installed - to upgrade all endpoints (that are providing services
towards Internet) at a large content site will take time, it might be
that the endpoints at the content site are upgraded only within their
normal life-cycle process. But if the size of the content site is
small the administrator either installs a CES-node or upgrades the
endpoints' stack - a choice to be taken and the decision will be
influenced by availability, reliability and economically feasibility.
Once the content sites has been upgraded the PI-address entries has
been removed from the DFZ. Most likely also some endpoints at the
consumer sites have been upgraded to support the hIPv4 stack -
especially if there have been issues with the caches or mapping
delays that have influenced on the service levels at the LCPs. Then,
how to keep track of the upgrade of the content sites - have they
been migrated or not? If the content sites or content endpoint has
been migrated the DNS records should have either a CES-node entry or
ALOC entry for each A-record. When the penetration of CES solutions
at content sites (followed up by CES-node/ALOC records in DNS) is
high enough the ISP can start to promote the hIPv4 stack upgrade at
the consumer sites. Once a PA-address block has been migrated it can
be released from global allocation to a regional allocation. Why
Frejborg Expires March 16, 2011 [Page 44]
Internet-Draft Hierarchical IPv4 Framework September 2010
would then an ISP push its customer to migrate to hIPv4 stacks? It is
due to the business model - it will be more expensive to stay in the
current architecture. The depletion of IPv4 addresses will either
cause more NAT at the service provider's network - operational
expenditures will increase because the network will become more
complex - or the ISP should force it's customers to migrate to IPv6 -
but the ISP could loose customers to other ISPs that are offering
IPv4 services. When PA-addresses has been migrated to the hIPv4
framework the ISP will have a more independent routing domain (ALOC
realm) with only ALOC prefixes from other ISPs and ELOC prefixes from
directly attached customers - BGP churn from other ISPs is no longer
received, also the amount of alternative paths is reduced and the ISP
can better control the growth of the RIB at their ALOC realm. The
operational and capital expenditures should be lower that than in the
current routing architecture.
To summarize, the content providers might find the CES+hIPv4 solution
attractive - it will remove the forthcoming IPv4 address depletion
constraints without forcing the consumer to switch to IPv6, thus the
content providers can continue to grow (reach more consumers). Also
the ISP might find this solution attractive; it should reduce the
capital and operational expenditures in long term. Both the content
providers and the ISPs are providing the foundation of Internet - if
both are adopting this architecture the consumers have to adopt, both
providers might find business models to "guide" the consumers towards
the new routing architecture.
Then how will this affect the consumer and content sites? The
residential users will need to upgrade their endpoints - but it
doesn't really matter what IP protocol version they use - it is the
availability and affordability of Internet that matters the most.
Enterprises will be affected a little bit more. The edge devices at
the enterprises local network needs to be upgraded - edge nodes such
as AS border routers, forwarding proxies, security nodes, remote
access service nodes, application delivery controllers, DNS, DHCP and
public nodes needs to be upgraded - but by installing a CES-node in
front of them, the upgrade process is postponed - instead the legacy
nodes can be upgraded during their normal life-cycle process. The
internal infrastructure is preserved as such, internal applications
can still use IPv4 and also all investment in IPv4 skills is
preserved.
Frejborg Expires March 16, 2011 [Page 45]
Internet-Draft Hierarchical IPv4 Framework September 2010
Walkthrough of use cases:
1. A legacy endpoint at a content site establishes a session to a
content site with a hIPv4 upgraded endpoint
When the legacy endpoint resolves the DNS entry for the remote
endpoint (a hIPv4 upgraded endpoint) the legacy endpoint will receive
an ALOC record in the DNS response. The legacy endpoint will ignore
the ALOC record, only the A-record is used to establish the session.
Next, the legacy endpoint will initialize the session and a packet is
sent towards the map-and-encapsulate ingress node, which need to do a
lookup at the CES mapping system (assumption here is that no cache
entry exist for the remote endpoint). The mapping system returns
either a CES-node prefix or an ALOC prefix for the lookup; since the
requested remote endpoint has been upgraded the mapping system
returns an ALOC prefix. The CES-node will not use the CES
encapsulation scheme for this session, instead the hIPv4 header
scheme shall be used and a /32 entry will be created in the cache. A
/32 entry must be created; it is possible that not all endpoints at
the remote site are upgraded to support the hIPv4 framework. The /32
cache entry can be replaced with a shorter prefix in the cache if all
endpoints are upgraded at the remote site. To indicate this situation
a sub-field shall be added for the ALOC record in the mapping system.
The CES-node must execute the following steps for the egress packets
a. verify IP- and transport header checksums
b. create the locator header, copy the destination address to the
ELOC field of the locator header
c. replace the destination address in the IP header with the ALOC
prefix given in the cache
d. insert the local CES-node prefix in the ALOC field of the
locator header
e. copy the transport protocol value of the IP header to the
protocol field of the locator header and set the hIPv4
protocol value in the protocol field of the IP header
f. set the desired parameters in the A-, I-, S-, FI-, VLB-, and
L-fields of the locator header
g. decrease TTL value with one
Frejborg Expires March 16, 2011 [Page 46]
Internet-Draft Hierarchical IPv4 Framework September 2010
h. calculate IP-, locator- and transport protocol header
checksums, transport protocol header calculation do not
include the locator header fields. When completed the packet
is transmitted
i. because the size of the packet might exceed MTU - due to
insertion of the locator header - if MTU is exceeded the CES-
node should inform the source endpoint with an ICMP message of
the situation and the CES-node should apply fragmentation of
the hIPv4 packet
2. A hIPv4 upgraded endpoint at a consumer/content site establishes a
session to a content site with a CES-node in front of a legacy
endpoint
The hIPv4 upgraded endpoint receives in the DNS response either an
ALOC record or a CES-node record for the resolved destination. From
the requesting hIPv4 endpoint's point of view it really doesn't
matter if the new record prefix is used to locate LSR-nodes or CES-
nodes in Internet - the CES-node will act as a hIPv4 proxy in front
of the remote legacy endpoint. Thus the hIPv4 endpoint assembles a
hIPv4 packet to initialize the session, when the packet arrives at
the CES-node it must execute the following:
a. verify the received packet that it uses the hIPv4 protocol
value in the protocol field of the IP header
b. verify IP-, locator- and transport protocol header checksums,
transport protocol header verification do not include the
locator header fields
c. replace the protocol field value of the IP header with the
protocol field value of the locator header
d. replace the source address in the IP header with the ELOC
prefix of the locator header
e. remove the locator header
f. create a cache entry (unless an entry already exists) for
returning packets, a /32 entry is required. To optimize the
usage of cache entries, the CES-node might ask the CES mapping
node if all endpoints at the remote site are upgraded or not -
if upgraded a shorter prefix can be used in the cache.
g. decrease TTL value with one
Frejborg Expires March 16, 2011 [Page 47]
Internet-Draft Hierarchical IPv4 Framework September 2010
h. calculate IP- and transport protocol header checksums
i. forward the packet upon the destination address of the
IP header
3. A hIPv4 enabled endpoint with a regional unique ELOC at a consumer
site establishes a session to a consumer site with a legacy endpoint.
In this use case the sessions will fail unless some mechanisms is
invented and implemented at the ISPs' map-and-encapsulate nodes. The
sessions will work inside an ALOC realm since the classical IPv4
framework is still valid, sessions between ALOC realms will fail.
Some applications are establishing sessions between consumer sites;
the most common are gaming and peer-to-peer applications. These
communities have historically been in the forefront to adopt new
technologies - it is expected that the affected communities either
develop workarounds to solve this issue or simply asking their
members to upgrade their stacks.
4. A legacy endpoint at a consumer/content site establishes a session
to a content site with a CES-node in front of a legacy endpoint
Assumed to be described in CES architecture drafts
5. A hIPv4 enabled endpoint at a consumer/content site establishes a
session to a content site with a hIPv4 enabled endpoint
See section 8
Author's Address
Patrick Frejborg
Email: pfrejborg@gmail.com
Frejborg Expires March 16, 2011 [Page 48]
