TEAS Working Group Daniele Ceccarelli (Ed)
Internet Draft Ericsson
Intended status: Informational Young Lee (Ed)
Expires: January 19, 2018 Huawei
July 20, 2017
Framework for Abstraction and Control of Traffic Engineered Networks
draft-ietf-teas-actn-framework-07
Abstract
Traffic Engineered networks have a variety of mechanisms to
facilitate the separation of the data plane and control plane. They
also have a range of management and provisioning protocols to
configure and activate network resources. These mechanisms
represent key technologies for enabling flexible and dynamic
networking.
Abstraction of network resources is a technique that can be applied
to a single network domain or across multiple domains to create a
single virtualized network that is under the control of a network
operator or the customer of the operator that actually owns
the network resources.
This document provides a framework for Abstraction and Control of
Traffic Engineered Networks (ACTN).
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
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 1]
Internet-Draft ACTN Framework July 2017
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
This Internet-Draft will expire on January 19, 2018.
Copyright Notice
Copyright (c) 2017 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.
Table of Contents
1. Introduction...................................................3
1.1. Terminology...............................................5
2. Business Model of ACTN.........................................9
2.1. Customers................................................10
2.2. Service Providers........................................10
2.3. Network Providers........................................12
3. Virtual Network Service.......................................12
4. ACTN Base Architecture........................................13
4.1. Customer Network Controller..............................15
4.2. Multi Domain Service Coordinator.........................16
4.3. Physical Network Controller..............................17
4.4. ACTN Interfaces..........................................17
5. Advanced ACTN architectures...................................18
5.1. MDSC Hierarchy for scalability...........................18
5.2. Functional Split of MDSC Functions in Orchestrators......19
6. Topology Abstraction Method...................................21
6.1. Abstraction Factors......................................22
6.2. Abstraction Types........................................23
6.2.1. Native/White Topology...............................23
6.2.2. Black Topology......................................24
6.2.3. Grey Topology.......................................25
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 2]
Internet-Draft ACTN Framework July 2017
6.3. Building Methods of Grey Topology........................27
6.3.1. Automatic generation of abstract topology by
configuration..............................................27
6.3.2. On-demand generation of supplementary topology via path
compute request/reply......................................28
6.4. Abstraction Configuration Consideration..................29
6.4.1. Packet Networks.....................................29
6.4.2. OTN Networks........................................29
6.4.3. WSON Networks.......................................30
6.5. Topology Abstraction Granularity Level example...........30
7. Access Points and Virtual Network Access Points...............32
7.1. Dual homing scenario.....................................34
8. Advanced ACTN Application: Multi-Destination Service..........35
8.1. Pre-Planned End Point Migration..........................36
8.2. On the Fly End Point Migration...........................37
9. Advanced Topic................................................37
10. Manageability Considerations.................................37
10.1. Policy..................................................38
10.2. Policy applied to the Customer Network Controller.......39
10.3. Policy applied to the Multi Domain Service Coordinator..39
10.4. Policy applied to the Physical Network Controller.......39
11. Security Considerations......................................40
11.1. Interface between the Customer Network Controller and Multi
Domain Service Coordinator (MDSC), CNC-MDSC Interface (CMI)...41
11.2. Interface between the Multi Domain Service Coordinator and
Physical Network Controller (PNC), MDSC-PNC Interface (MPI)...41
12. References...................................................42
12.1. Informative References..................................42
13. Contributors.................................................43
Authors' Addresses...............................................44
APPENDIX A - Example of MDSC and PNC functions integrated in
Service/Network Orchestrator.....................................45
APPENDIX B - Example of IP + Optical network with L3VPN service..45
1. Introduction
Traffic Engineered networks have a variety of mechanisms to
facilitate separation of data plane and control plane including
distributed signaling for path setup and protection, centralized
path computation for planning and traffic engineering, and a range
of management and provisioning protocols to configure and activate
network resources. These mechanisms represent key technologies for
enabling flexible and dynamic networking.
The term Traffic Engineered network is used in this document to
refer to a network that uses any connection-oriented technology
under the control of a distributed or centralized control plane to
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 3]
Internet-Draft ACTN Framework July 2017
support dynamic provisioning of end-to-end connectivity. Some
examples of networks that are in scope of this definition are
optical networks, MPLS Transport Profile (MPLS-TP) networks
[RFC5654], and MPLS Traffic Engineering (MPLS-TE) networks
[RFC2702].
One of the main drivers for Software Defined Networking (SDN)
[RFC7149] is a decoupling of the network control plane from the data
plane. This separation of the control plane from the data plane has
been already achieved with the development of MPLS/GMPLS [GMPLS] and
the Path Computation Element (PCE) [RFC4655] for TE-based networks.
One of the advantages of SDN is its logically centralized control
regime that allows a global view of the underlying networks.
Centralized control in SDN helps improve network resource
utilization compared with distributed network control. For TE-based
networks, PCE is essentially equivalent to a logically centralized
path computation function.
Three key aspects that need to be solved by SDN are:
. Separation of service requests from service delivery so that
the orchestration of a network is transparent from the point of
view of the customer but remains responsive to the customer's
services and business needs.
. Network abstraction: As described in [RFC7926], abstraction is
the process of applying policy to a set of information about a
TE network to produce selective information that represents the
potential ability to connect across the domain. The process of
abstraction presents the connectivity graph in a way that is
independent of the underlying network technologies,
capabilities, and topology so that it can be used to plan and
deliver network services in a uniform way
. Coordination of resources across multiple domains and multiple
layers to provide end-to-end services regardless of whether the
domains use SDN or not.
As networks evolve, the need to provide separated service
request/orchestration and resource abstraction has emerged as a key
requirement for operators. In order to support multiple clients each
with its own view of and control of the server network, a network
operator needs to partition (or "slice") the network resources. The
resulting slices can be assigned to each client for guaranteed usage
which is a step further than shared use of common network resources.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 4]
Internet-Draft ACTN Framework July 2017
Furthermore, each network represented to a client can be built from
abstractions of the underlying networks so that, for example, a link
in the client's network is constructed from a path or collection of
paths in the underlying network.
We call the set of management and control functions used to provide
these features Abstraction and Control of Traffic Engineered
Networks (ACTN).
Particular attention needs to be paid to the multi-domain case, ACTN
can facilitate virtual network operation via the creation of a
single virtualized network or a seamless service. This supports
operators in viewing and controlling different domains (at any
dimension: applied technology, administrative zones, or vendor-
specific technology islands) as a single virtualized network.
The ACTN framework described in this document facilitates:
. Abstraction of the underlying network resources to higher-layer
applications and customers [RFC7926].
. Virtualization of particular underlying resources, whose
selection criterion is the allocation of those resources to a
particular customer, application or service [ONF-ARCH].
. Network slicing of infrastructure to meet specific customers'
service requirements.
. Creation of a virtualized environment allowing operators to
view and control multi-domain networks as a single virtualized
network.
. The presentation to customers of networks as a virtual network
via open and programmable interfaces.
1.1. Terminology
The following terms are used in this document. Some of them are
newly defined, some others reference existing definition:
. Network Slicing: In the context of ACTN, network slicing is a
collection of resources that are used to establish logically
dedicated virtual networks over TE networks. It allows a
network provider to provide dedicated virtual networks for
application/customer over a common network infrastructure. The
logically dedicated resources are a part of the larger common
network infrastructures that are shared among various network
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 5]
Internet-Draft ACTN Framework July 2017
slice instances which are the end-to-end realization of network
slicing, consisting of the combination of physically or
logically dedicated resources.
. Node: A node is a vertex on the graph representation of a TE
topology. In a physical network topology, a node corresponds to
a physical network element (NE). In an abstract network
topology, a node (sometimes called an abstract node) is a
representation as a single vertex of one or more physical NEs
and their connecting physical connections. The concept of a
node represents the ability to connect from any access to the
node (a link end) to any other access to that node, although
"limited cross-connect capabilities" may also be defined to
restrict this functionality. Just as network slicing and
network abstraction may be applied recursively, so a node in a
topology may be created by applying slicing or abstraction on
the nodes in the underlying topology.
. Link: A link is an edge on the graph representation of a TE
topology. Two nodes connected by a link are said to be
"adjacent" in the TE topology. In a physical network topology,
a link corresponds to a physical connection. In an abstract
network topology, a link (sometimes called an abstract link) is
a representation of the potential to connect a pair of points
with certain TE parameters (see RFC 7926 for details). Network
slicing/virtualization and network abstraction may be applied
recursively, so a link in a topology may be created by applying
slicing and/or abstraction on the links in the underlying
topology.
. CNC: A Customer Network Controller is responsible for
communicating customer's virtual network service requirements
to network provider. It has knowledge of the end-point
associated with virtual network service, service policy, and
other QoS information related to the service it is responsible
for instantiating.
. PNC: A Physical Network Controller is responsible for
controlling devices or NEs under its direct control. The PNC
functions can be implemented as part of an SDN domain
controller, a Network Management System (NMS), an Element
Management System (EMS), an active PCE-based controller or any
other means to dynamically control a set of nodes and that is
implementing an NBI compliant with ACTN specification.
. PNC domain: A PNC domain includes all the resources under the
control of a single PNC. It can be composed of different
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 6]
Internet-Draft ACTN Framework July 2017
routing domains and administrative domains, and the resources
may come from different layers. The interconnection between PNC
domains can be a link or a node.
_______ Border Link _______
_( )================( )_
_( )_ _( )_
( ) ---- ( )
( PNC )| |( PNC )
( Domain X )| |( Domain Y )
( )| |( )
(_ _) ---- (_ _)
(_ _) Border (_ _)
(_______) Node (_______)
Figure 1: PNC Domain Borders
. MDSC: A multi-domain Service Coordinator is a functional block
that implements all four ACTN main functions, i.e., multi
domain coordination, virtualization/abstraction, customer
mapping/translation, and virtual service coordination. The
first two functions of the MDSC, namely, multi domain
coordination and virtualization/abstraction are referred to as
network-related functions while the last two functions, namely,
customer mapping/translation and virtual service coordination
are referred to as service-related functions. See details on
these functions in Section 4.2. In some implementation, PNC and
MDSC functions can be co-located and implemented in the same
box.
. A Virtual Network (VN) is a customer view of the TE
network. Depending on the agreement between client and
provider various VN operations and VN views are possible as
follows:
o VN Creation - VN could be pre-configured and created via
offline negotiation between customer and provider. In
other cases, the VN could also be created dynamically
based on a request from the customer with given SLA
attributes which satisfy the customer's objectives.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 7]
Internet-Draft ACTN Framework July 2017
o Dynamic Operations - The VN could be further modified or
deleted based on a customer request. The customer can
further act upon the virtual network resources to perform
end-to-end tunnel management (set-up/release/modify).
These changes will result in subsequent LSP management at
the operator's level.
o VN Type:
a. The VN can be seen as set of end-to-end tunnels from a
customer point of view, where each tunnel is referred
as a VN member. Each VN member can then be formed by
recursive slicing or abstraction of paths in
underlying networks. Such end-to-end tunnels may
comprise of customer end points, access links, intra-
domain paths, and inter-domain links. In this view, VN
is thus a set of VN members (which is referred to as
Type 1 VN)
b. The VN can also be seen as a topology comprising of
physical, sliced, and abstract nodes and links. This
VN is referred to as Type 2 VN. The nodes in this case
include physical customer end points, border nodes,
and internal nodes as well as abstracted nodes.
Similarly the links include physical access links,
inter-domain links, and intra-domain links as well as
abstract links. With VN type 2, it is still possible
to view VN member-level.
. Virtual Network Service (VNS) is requested by the customer and
negotiated with the provider. There are three types of VNS
defined in this document. Type 1 VNS refers to VNS in which
customer is allowed to create and operate a Type 1 VN. Type 2a
and 2b VNS refers to the VNS in which customer is allowed to
create and operates a Type 2 VN. With Type 2a VNS, once the VN
is statically created at service configuration time, the
customer is not allowed to change the topology (i.e., adding or
deleting abstract nodes/links). Type 2b VNS is the same as Type
2a VNS except that the customer is allowed to change topology
dynamically from the initial topology created at service
configuration time. See Section 3 for details.
. Abstraction. This process is defined in [RFC7926].
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 8]
Internet-Draft ACTN Framework July 2017
. Abstract Link: The term "abstract link" is defined in
[RFC7926].
. Abstract Topology: The topology of abstract nodes and abstract
links presented through the process of abstraction by a lower
layer network for use by a higher layer network.
. Access link: A link between a customer node and a provider
node.
. Inter-domain link: A link between domains managed by different
PNCs. The MDSC is in charge of managing inter-domain links.
. Access Point (AP): An access point is used to keep
confidentiality between the customer and the provider. It is a
logical identifier shared between the customer and the
provider, used to map the end points of the border node in both
the customer and the provider NW. The AP can be used by the
customer when requesting VN service to the provider.
. VN Access Point (VNAP): A VNAP is defined as the binding
between an AP and a given VN and is used to identify the
portion of the access and/or inter-domain link dedicated to a
given VN.
2. Business Model of ACTN
The Virtual Private Network (VPN) [RFC4026] and Overlay Network (ON)
models [RFC4208] are built on the premise that the network provider
provides all virtual private or overlay networks to its customers.
These models are simple to operate but have some disadvantages in
accommodating the increasing need for flexible and dynamic network
virtualization capabilities.
There are three key entities in the ACTN model:
- Customers
- Service Providers
- Network Providers
These are described in the following sections.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 9]
Internet-Draft ACTN Framework July 2017
2.1. Customers
Within the ACTN framework, different types of customers may be taken
into account depending on the type of their resource needs, and on
their number and type of access. For example, it is possible to
group them into two main categories:
Basic Customer: Basic customers include fixed residential users,
mobile users and small enterprises. Usually, the number of basic
customers for a service provider is high: they require small amounts
of resources and are characterized by steady requests (relatively
time invariant). A typical request for a basic customer is for a
bundle of voice services and internet access. Moreover, basic
customers do not modify their services themselves: if a service
change is needed, it is performed by the provider as a proxy and the
services generally have very few dedicated resources (such as for
subscriber drop), with everything else shared on the basis of some
Service Level Agreement (LSA), which is usually best-efforts.
Advanced Customer: Advanced customers typically include enterprises,
governments and utilities. Such customers can ask for both point-to
point and multipoint connectivity with high resource demands varying
significantly in time and from customer to customer. This is one of
the reasons why a bundled service offering is not enough and it is
desirable to provide each advanced customer with a customized
virtual network service.
Advanced customers may own dedicated virtual resources, or share
resources. They may also have the ability to modify their service
parameters within the scope of their virtualized environments. The
primary focus of ACTN is Advanced Customers.
As customers are geographically spread over multiple network
provider domains, they have to interface to multiple providers and
may have to support multiple virtual network services with different
underlying objectives set by the network providers. To enable these
customers to support flexible and dynamic applications they need to
control their allocated virtual network resources in a dynamic
fashion, and that means that they need a view of the topology that
spans all of the network providers. Customers of a given service
provider can in turn offer a service to other customers in a
recursive way.
2.2. Service Providers
Service providers are the providers of virtual network services (see
Section 3 for details) to their customers. Service providers may or
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 10]
Internet-Draft ACTN Framework July 2017
may not own physical network resources (i.e, may or may not be
network providers as described in Section 2.3). When a service
provider is the same as the network provider, this is similar to
existing VPN models applied to a single provider. This approach
works well when the customer maintains a single interface with a
single provider. When customer spans multiple independent network
provider domains, then it becomes hard to facilitate the creation of
end-to-end virtual network services with this model.
A more interesting case arises when network providers only provide
infrastructure, while distinct service providers interface to the
customers. In this case, service providers are, themselves customers
of the network infrastructure providers. One service provider may
need to keep multiple independent network providers as its end-users
span geographically across multiple network provider domains.
The ACTN network model is predicated upon this three tier model and
is summarized in Figure 2:
+----------------------+
| customer |
+----------------------+
|
VNS || | /\ VNS
Request || | || Reply
\/ | ||
+----------------------+
| Service Provider |
+----------------------+
/ | \
/ | \
/ | \
/ | \
+------------------+ +------------------+ +------------------+
|Network Provider 1| |Network Provider 2| |Network Provider 3|
+------------------+ +------------------+ +------------------+
Figure 2: Three tier model.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 11]
Internet-Draft ACTN Framework July 2017
There can be multiple service providers to which a customer may
interface.
There are multiple types of service providers, for example:
. Data Center providers can be viewed as a service provider type
as they own and operate data center resources for various WAN
customers, and they can lease physical network resources from
network providers.
. Internet Service Providers (ISP) are service providers of
internet services to their customers while leasing physical
network resources from network providers.
. Mobile Virtual Network Operators (MVNO) provide mobile services
to their end-users without owning the physical network
infrastructure.
2.3. Network Providers
Network Providers are the infrastructure providers that own the
physical network resources and provide network resources to their
customers. The layered model described in this architecture
separates the concerns of network providers and customers, with
service providers acting as aggregators of customer requests.
3. Virtual Network Service
Virtual Network Service (VNS) is requested by the customer and
negotiated with the provider. There are three types of VNS defined
in this document.
Type 1 VNS refers to VNS in which customer is allowed to create and
operate a Type 1 VN. Type 1 VN is a VN that comprises a set of end-
to-end tunnels from a customer point of view, where each tunnel is
referred as a VN member. With Type 1 VNS, the network operator does
not need to provide additional abstract VN topology associated with
the Type 1 VN.
Type 2a VNS refer to VNS in which customer is allowed to create and
operates a Type 2 VN, but not allowed to change topology once it is
configured at service configuration time. Type 2 VN is an abstract
VN topology that may comprise of virtual/abstract nodes and links.
The nodes in this case may include physical customer end points,
border nodes, and internal nodes as well as abstracted nodes.
Similarly, the links may include physical access links, inter-domain
links, and intra-domain links as well as abstract links.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 12]
Internet-Draft ACTN Framework July 2017
Type 2b VNS refers to VNS in which customer is allowed to create and
operate a Type 2 VN and the customer is allowed to dynamically
change abstract VN topology from the initially configured abstract
VN topology at service configuration time.
From an implementation standpoint, Type 2a VNS and Type 2b VNS
differentiation might be fulfilled via local policy.
In all types of VNS, customer can specify a set of service related
parameters such as connectivity type, VN traffic matrix (e.g.,
bandwidth, latency, diversity, etc.), VN survivability, VN service
policy and other characteristics.
4. ACTN Base Architecture
This section provides a high-level model of ACTN showing the
interfaces and the flow of control between components.
The ACTN architecture is aligned with the ONF SDN architecture [ONF-
ARCH] and presents a 3-tiers reference model. It allows for
hierarchy and recursiveness not only of SDN controllers but also of
traditionally controlled domains that use a control plane. It
defines three types of controllers depending on the functionalities
they implement. The main functionalities that are identified are:
. Multi-domain coordination function: This function oversees the
specific aspects of the different domains and builds a single
abstracted end-to-end network topology in order to coordinate
end-to-end path computation and path/service provisioning.
Domain sequence path calculation/determination is also a part
of this function.
. Virtualization/Abstraction function: This function provides an
abstracted view of the underlying network resources for use by
the customer - a customer may be the client or a higher level
controller entity. This function includes network path
computation based on customer service connectivity request
constraints, path computation based on the global network-wide
abstracted topology, and the creation of an abstracted view of
network resources allocated to each customer. These operations
depend on customer-specific network objective functions and
customer traffic profiles.
. Customer mapping/translation function: This function is to map
customer requests/commands into network provisioning requests
that can be sent to the Physical Network Controller (PNC)
according to business policies provisioned statically or
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 13]
Internet-Draft ACTN Framework July 2017
dynamically at the OSS/NMS. Specifically, it provides mapping and
translation of a customer's service request into a set of
parameters that are specific to a network type and technology
such that network configuration process is made possible.
. Virtual service coordination function: This function translates
customer service-related information into virtual network
service operations in order to seamlessly operate virtual
networks while meeting a customer's service requirements. In
the context of ACTN, service/virtual service coordination
includes a number of service orchestration functions such as
multi-destination load balancing, guarantees of service
quality, bandwidth and throughput. It also includes
notifications for service fault and performance degradation and
so forth.
Figure 3 depicts the base ACTN architecture with three controller
types and the corresponding interfaces between these controllers.
The types of controller defined in the ACTN architecture are shown
in Figure 3 below and are as follows:
. CNC - Customer Network Controller
. MDSC - Multi Domain Service Coordinator
. PNC - Physical Network Controller
Figure 3 also shows the following interfaces:
. CMI - CNC-MDSC Interface
. MPI - MDSC-PNC Interface
. SBI - South Bound Interface
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 14]
Internet-Draft ACTN Framework July 2017
+--------------+ +---------------+ +--------------+
| CNC-A | | CNC-B | | CNC-C |
|(DC provider) | | (ISP) | | (MVNO) |
+--------------+ +---------------+ +--------------+
\ | /
Business \ | /
Boundary =======\========================|=========================/=======
Between \ | CMI /
Customer & ----------- | --------------
Network Provider \ | /
+-----------------------+
| MDSC |
+-----------------------+
/ | \
------------ |MPI ----------------
/ | \
+-------+ +-------+ +-------+
| PNC | | PNC | | PNC |
+-------+ +-------+ +-------+
| GMPLS / | / \
| trigger / |SBI / \
-------- ----- | / \
( ) ( ) | / \
- - ( Phys. ) | / -----
( GMPLS ) ( Net ) | / ( )
( Physical ) ---- | / ( Phys. )
( Network ) ----- ----- ( Net )
- - ( ) ( ) -----
( ) ( Phys. ) ( Phys. )
-------- ( Net ) ( Net )
----- -----
Figure 3: ACTN Base Architecture
4.1. Customer Network Controller
A Virtual Network Service is instantiated by the Customer Network
Controller via the CNC-MDSC Interface (CMI). As the Customer Network
Controller directly interfaces to the applications, it understands
multiple application requirements and their service needs. It is
assumed that the Customer Network Controller and the MDSC have a
common knowledge of the end-point interfaces based on their business
negotiations prior to service instantiation. End-point interfaces
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 15]
Internet-Draft ACTN Framework July 2017
refer to customer-network physical interfaces that connect customer
premise equipment to network provider equipment.
4.2. Multi Domain Service Coordinator
The Multi Domain Service Coordinator (MDSC) sits between the CNC
that issues connectivity requests and the Physical Network
Controllers (PNCs) that manage the physical network resources. The
MDSC can be collocated with the PNC.
The internal system architecture and building blocks of the MDSC are
out of the scope of ACTN. Some examples can be found in the
Application Based Network Operations (ABNO) architecture [RFC7491]
and the ONF SDN architecture [ONF-ARCH].
The MDSC is the only building block of the architecture that can
implement all four ACTN main functions, i.e., multi domain
coordination, virtualization/abstraction, customer
mapping/translation, and virtual service coordination. The first two
functions of the MDSC, namely, multi domain coordination and
virtualization/abstraction are referred to as network-related
functions while the last two functions, namely, customer
mapping/translation and virtual service coordination are referred to
as service-related functions.
The key point of the MDSC (and of the whole ACTN framework) is
detaching the network and service control from underlying technology
to help the customer express the network as desired by business
needs. The MDSC envelopes the instantiation of the right technology
and network control to meet business criteria. In essence it
controls and manages the primitives to achieve functionalities as
desired by the CNC.
In order to allow for multi-domain coordination a 1:N relationship
must be allowed between MDSCs and between MDSCs and PNCs (i.e. 1
parent MDSC and N child MDSC or 1 MDSC and N PNCs).
In addition to that, it could also be possible to have an M:1
relationship between MDSCs and PNC to allow for network resource
partitioning/sharing among different customers not necessarily
connected to the same MDSC (e.g., different service providers).
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 16]
Internet-Draft ACTN Framework July 2017
4.3. Physical Network Controller
The Physical Network Controller (PNC) oversees configuring the
network elements, monitoring the topology (physical or virtual) of
the network, and passing information about the topology (either raw
or abstracted) to the MDSC.
The internal architecture of the PNC, its building blocks, and the
way it controls its domain are out of the scope of ACTN. Some
examples can be found in the Application Based Network Operations
(ABNO) architecture [RFC7491] and the ONF SDN architecture [ONF-
ARCH]
The PNC, in addition to being in charge of controlling the physical
network, is able to implement two of the four main ACTN main
functions: multi domain coordination and virtualization/abstraction
function.
Note that from an implementation point of view it is possible to
integrate one or more MDSC functions and one or more PNC functions
within the same controller.
4.4. ACTN Interfaces
The network has to provide open, programmable interfaces, through
which customer applications can create, replace and modify virtual
network resources and services in an interactive, flexible and
dynamic fashion while having no impact on other customers. Direct
customer control of transport network elements and virtualized
services is not perceived as a viable proposition for transport
network providers due to security and policy concerns among other
reasons. In addition, the network control plane for transport
networks has been separated from the data plane and as such it is
not viable for the customer to directly interface with transport
network elements.
. CMI Interface: The CNC-MDSC Interface (CMI) is an interface
between a CNC and an MDSC. As depicted in Figure 3, the CMI is
a business boundary between customer and network provider. It
is used to request virtual network services required for the
applications. Note that all service related information such as
specific service properties, including virtual network service
type, topology, bandwidth, and constraint information are
conveyed over this interface. Most of the information over this
interface is technology agnostic; however, there are some
cases, e.g., access link configuration, where it should be
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 17]
Internet-Draft ACTN Framework July 2017
possible to explicitly request for a VN to be created at a
given layer in the network (e.g. ODU VN or MPLS VN).
. MPI Interface: The MDSC-PNC Interface (MPI) is an interface
between an MDSC and a PNC. It communicates the creation
requests for new connectivity or for bandwidth changes in the
physical network. In multi-domain environments, the MDSC needs
to establish multiple MPIs, one for each PNC, as there is one
PNC responsible for control of each domain. The MPI could have
different degrees of abstraction and present an abstracted
topology hiding technology specific aspects of the network or
convey technology specific parameters to allow for path
computation at the MDSC level. Please refer to CCAMP Transport
NBI work for the latter case [Transport NBI].
. SBI Interface: This interface is out of the scope of ACTN. It
is shown in Figure 3 for reference reason only.
Please note that for all the three interfaces, when technology
specific information needs to be included, this info will be add-ons
on top of the general abstract topology. As far as general topology
abstraction standpoint, all interfaces are still recursive in
nature.
5. Advanced ACTN architectures
This section describes advanced forms of ACTN architectures as
possible implementation choices.
5.1. MDSC Hierarchy for scalability
A hierarchy of MDSCs can be foreseen for many reasons, among which
are scalability, administrative choices or putting together
different layers and technologies in the network. In the case where
there is a hierarchy of MDSCs, we introduce the higher-level MDSC
(MDSC-H) the lower-level MDSC (MDSC-L) and the interface between
them is basically of a recursive nature of the MPI. An
implementation choice could foresee the usage of an MDSC-L for all
the PNCs related to a given network layer or technology (e.g.
IP/MPLS) a different MDSC-L for the PNCs related to another
layer/technology (e.g. OTN/WDM) and an MDSC-H to coordinate them.
Figure 4 shows this case.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 18]
Internet-Draft ACTN Framework July 2017
+--------+
| CNC |
+--------+
|
|
+----------+
--------| MDSC-H |--------
| +----------+ |
| |
+---------+ +---------+
| MDSC-L | | MDSC-L |
+---------+ +---------+
Figure 4: MDSC Hierarchy
Note that both the MDSC-H and the MDSC-L in general cases implement
all four functions of the MDSC discussed in Section 3.2.
5.2. Functional Split of MDSC Functions in Orchestrators
Another implementation choice could foresee the separation of MDSC
functions into two groups (i.e., one group for service-related
functions and another group for network-related functions) which
will result in a service orchestrator for providing service-related
functions of MDSC and other non-ACTN functions and a network
orchestrator for providing network-related functions of MDSC and
other non-ACTN functions. Figure 5 shows this case and it also
depicts the mapping between ACTN architecture and the YANG service
model architecture described in [Service-YANG]. This mapping is
helpful for the readers who are not familiar with some TEAS specific
terminology used in this document. A number of key ACTN interfaces
exist for deployment and operation of ACTN-based networks. These are
highlighted in Figure 5 (ACTN Interfaces).
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 19]
Internet-Draft ACTN Framework July 2017
+------------------------------+
| Customer |
| +-----+ +----------+ |
| | CNC | |Other fns.| |
| +-----+ +----------+ |
+------------------------------+
| Customer Service Model
|
+-----------------------------------------------+
********|********************** Service Orchestrator |
* MDSC | +------+ +------+ * +-----------+ |
* | | MDSC | | MDSC | * | Other fns.| |
* | | F1 | | F2 | * | (non-ACTN)| |
* | +------+ +------+ * +-----------+ |
* +---------------------*-------------------------+
* * | Service Delivery Model
* * |
* +---------------------*-------------------------+
* | * Network Orchestrator |
* | +------+ +------+ * +-----------+ |
* | | MDSC | | MDSC | * | Other fns.| |
* | | F3 | | F4 | * | (non-ACTN)| |
* | +------+ +------+ * +-----------+ |
********|********************** |
+-----------------------------------------------+
| Network Configuration Model
|
+-------------------------------------------+
| Domain Controller |
| +------+ +-----------+ |
| | PNC | | Other fns.| |
| +------+ | (non-ACTN)| |
| +-----------+ |
+-------------------------------------------+
| Device Configuration Model
|
--------
| Device |
--------
Figure 5: ACTN Architecture in the context of YANG Service Models
In Figure 5, MDSC F1 and F2 correspond to customer
mapping/translation, and virtual service coordination, respectively,
which are the MDSC service-related functions as defined in Section
4. MDSC F3 and F4 correspond to multi domain coordination,
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 20]
Internet-Draft ACTN Framework July 2017
virtualization/abstraction, respectively, which are the MDSC
network-related functions as defined in Section 4. In some
implementation, MDSC F1 and F2 can be implemented as part of a
Service Orchestrator which may support other non-ACTN functions.
Likewise, the MDSC F3 and F4 can be implemented as part of a Network
Orchestrator which may support other non-ACTN MDSC functions.
Also note that the PNC is not same as domain controller. Domain
controller in general has a larger set of functions than that of
PNC. The main functions of PNC are explained in Section 3.3.
Likewise, Customer has a larger set of functions than that of the
CNC.
Customer service model describes a service as offer or delivered to
a customer by a network operator as defined in [Service-YANG]. The
CMI is a subset of a customer service model to support VNS. This
model encompasses other non-TE/non-ACTN models to control non-ACTN
services (e.g., L3SM).
Service delivery model is used by a network operator to define and
configure how a service is provided by the network as defined in
[Service-YANG]. This model is similar to the MPI model as the
network-related functions of the MDSC, i.e., F3 and F4, provide an
abstract topology view of the E2E network to the service-related
functions of the MDSC, i.e., F1 and F2, which translate customer's
request at the CMI into the network configuration at the MPI.
Network configuration model is used by a network orchestrator to
provide network-level configuration model to a controller as defined
in [Service-YANG]. The MPI is a subset of network configuration
model to support TE configuration. This model encompasses the MPI
model plus other non-TE/non-ACTN models to control non-ACTN
functions of the domain controller (e.g., L3VPN).
Device configuration model is used by a controller to configure
physical network elements.
6. Topology Abstraction Method
This section discusses topology abstraction factors, types and their
context in ACTN architecture. Topology abstraction is useful in ACTN
architecture as a way to scale multi-domain network operation. Note
that this is the abstraction performed by the PNC to the MDSC or by
the MDSC-L to the MDSC-H, and that this is different from the VN
Type 2 topology (that is created and negotiated between the CNC and
the MDSC as part of the VNS). The purpose of topology abstraction
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 21]
Internet-Draft ACTN Framework July 2017
discussed in this section is for an efficient internal network
operation based on abstraction principle.
6.1. Abstraction Factors
This section provides abstraction factors in the ACTN architecture.
The MDSC oversees the specific aspects of the different domains and
builds a single abstracted end-to-end network topology in order to
coordinate end-to-end path computation and path/service
provisioning. In order for the MDSC to perform its coordination
function, it depends on the coordination with the PNCs which are the
domain-level controllers especially as to what level of domain
network resource abstraction is agreed upon between the MDSC and the
PNCs.
As discussed in [RFC7926], abstraction is tied with policy of the
networks. For instance, per an operational policy, the PNC would not
be allowed to provide any technology specific details (e.g., optical
parameters for WSON) in its update. In such case, the abstraction
level of the update will be in a generic nature. In order for the
MDSC to get technology specific topology information from the PNC, a
request/reply mechanism may be employed.
In some cases, abstraction is also tied with the controller's
capability of abstraction as it involves some rules and algorithms
to be applied to the actual network resource information (which is
also known as network topology).
[TE-Topology] describes YANG models for TE-network abstraction.
[PCEP-LS] describes PCEP Link-state mechanism that also allows for
transport of abstract topology in the context of Hierarchical PCE.
There are factors that may impact the choice of abstraction. Here
are the most relevant:
- The nature of underlying domain networks: Abstraction depends on
the nature of the underlying domain networks. For instance, packet
networks may have different level of abstraction requirements from
that of optical networks. Within optical networks, WSON may have
different level of abstraction requirements than the OTN networks.
- The capability of the PNC: Abstraction depends on the capability
of the PNCs. As abstraction requires hiding details of the
underlying resource network resource information, the PNC
capability to run some internal optimization algorithm impacts the
feasibility of abstraction. Some PNC may not have the ability to
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 22]
Internet-Draft ACTN Framework July 2017
abstract native topology while other PNCs may have such an ability
to abstract actual topology by using sophisticated algorithms.
- Scalability factor: Abstraction is a function of scalability. If
the actual network resource information is of small size, then the
need for abstraction would be less than the case where the native
network resource information is of large size. In some cases,
abstraction may not be needed at all.
- The frequency of topology updates: The proper abstraction level
may depend on the frequency of topology updates and vice versa.
- The capability/nature of the MDSC: The nature of the MDSC impacts
the degree/level of abstraction. If the MDSC is not capable of
handling optical parameters such as those specific to OTN/WSON,
then white topology abstraction may not work well.
- The confidentiality: In some cases where the PNC would like to
hide key internal topological data from the MDSC, the abstraction
method should consider this aspect.
- The scope of abstraction: All of the aforementioned factors are
equally applicable to both the MPI (MDSC-PNC Interface) and the
CMI (CNC-MDSC Interface).
6.2. Abstraction Types
This section defines the following three types of topology
abstraction:
. Native/White Topology (Section 6.2.1)
. Black Topology (Section 6.2.2)
. Grey Topology (Section 6.2.3)
6.2.1. Native/White Topology
This is a case where the PNC provides the actual network topology to
the MDSC without any hiding or filtering of information as shown in
Figure 6a. In this case, the MDSC has the full knowledge of the
underlying network topology and as such there is no need for the
MDSC to send a path computation request to the PNC. The computation
burden will fall on the MDSC to find an optimal end-to-end path and
optimal per domain paths.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 23]
Internet-Draft ACTN Framework July 2017
+--+ +--+ +--+ +--+
+-+ +-----+ +-----+ +-----+ +-+
++-+ ++-+ +-++ +-++
| | | |
| | | |
| | | |
| | | |
++-+ ++-+ +-++ +-++
+-+ +-----+ +-----+ +-----+ +-+
+--+ +--+ +--+ +--+
Figure 6a: The native/white topology
6.2.2. Black Topology
The entire domain network is abstracted as a single virtual node
(see the definition of virtual node in [RFC7926]) with the
access/egress links without disclosing any node internal
connectivity information.
Figure 6b depicts a native topology with the corresponding black
topology with one virtual node and inter-domain links. In this case,
the MDSC has to make path computation requests to the PNCs before it
can determine an end-to-end path. If there are a large number of
inter-connected domains, this abstraction method may impose a heavy
coordination load at the MDSC level in order to find an optimal end-
to-end path.
The black topology would not give the MDSC any critical network
resource information other than the border nodes/links information
and as such it is likely to have a need for complementary
communications between the MDSC and the PNCs (e.g., Path computation
Request/Reply).
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 24]
Internet-Draft ACTN Framework July 2017
+--+ +--+ +--+ +--+
+-+ +-----+ +-----+ +-----+ +-+
++-+ ++-+ +-++ +-++
| | | |
| | | |
| | | |
| | | |
++-+ ++-+ +-++ +-++
+-+ +-----+ +-----+ +-----+ +-+
+--+ +--+ +--+ +--+
+--------+
+--+ +--+
| |
| |
| |
| |
| |
| |
+--+ +--+
+--------+
Figure 6b: The native topology and the corresponding black topology
with one virtual node and inter-domain links
6.2.3. Grey Topology
This abstraction level, referred to a grey topology, represents a
compromise between black and white topology from a granularity point
of view. As shown in Figures 7a and 7b, we may further differentiate
from a perspective of how to abstract internal TE resources between
the pairs of border nodes:
. Grey topology type A: border nodes with a TE links between them
in a full mesh fashion (See Figure 7a).
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 25]
Internet-Draft ACTN Framework July 2017
+--+ +--+ +--+ +--+
+-+ +-----+ +-----+ +-----+ +-+
++-+ ++-+ +-++ +-++
| | | |
| | | |
| | | |
| | | |
++-+ ++-+ +-++ +-++
+-+ +-----+ +-----+ +-----+ +-+
+--+ +--+ +--+ +--+
+--+ +--+
+-+ +----+ +-+
++-+ +-++
| \ / |
| \/ |
| /\ |
| / \ |
++-+ +-++
+-+ +----+ +-+
+--+ +--+
Figure 7a: The native topology and the corresponding grey topology
type A with TE links between border nodes
For each pair of ingress and egress nodes (i.e., border nodes
to/from the domain), TE link metric is provided with TE attributes
such as max bandwidth available, link delay, etc. This abstraction
depends on the underlying TE networks.
Note that this grey topology can also be represented as a single
abstract node with the connectivity matrix defined in [TE-Topology],
abstracting the internal connectivity information. The only thing
might be different is some additional information about the end
points of the links of the border nodes (i.e., links outward
customer-facing) as they cannot be included in the connectivity
matrix's termination points.
. Grey topology type B: border nodes with some internal
abstracted nodes and abstracted links (See Figure 7b)
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 26]
Internet-Draft ACTN Framework July 2017
+--+ +--+ +--+
+-+ +-----+ +-----+ +-+
++-+ ++-+ +-++
| |
| |
| |
| |
++-+ ++-+ +-++
+-+ +-----+ +-----+ +-+
+--+ +--+ +--+
Figure 7b: The grey topology type B with abstract nodes/links
between border nodes
The grey abstraction type B would allow the MDSC to have more
information about the internals of the domain networks by the PNCs
so that the MDSC can flexibly determine optimal paths. The MDSC may
configure some of the internal virtual nodes (e.g., cross-connect)
to redirect its traffic as it sees changes from the domain networks.
6.3. Building Methods of Grey Topology
This section discusses two different methods of building a grey
topology:
. Automatic generation of abstract topology by configuration
(Section 6.3.1)
. On-demand generation of supplementary topology via path
computation request/reply (Section 6.3.2)
6.3.1. Automatic generation of abstract topology by configuration
The "Automatic generation" method is based on the
abstraction/summarization of the whole domain by the PNC and its
advertisement on MPI interface once the abstraction level is
configured. The level of abstraction advertisement can be decided
based on some PNC configuration parameters (e.g. provide the
potential connectivity between any PE and any ASBR in an MPLS-TE
network.
Note that the configuration parameters for this potential topology
can include available B/W, latency, or any combination of defined
parameters. How to generate such tunnel information is beyond the
scope of this document.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 27]
Internet-Draft ACTN Framework July 2017
Such potential topology needs to be periodically or
incrementally/asynchronously updated every time that a failure, a
recovery or the setup of new VNs causes a change in the
characteristics of the advertised grey topology (e.g. in our
previous case if due to changes in the network is it now possible to
provide connectivity between a given PE and a given ASBR with a
higher delay in the update).
6.3.2. On-demand generation of supplementary topology via path compute
request/reply
The "on-demand generation" of supplementary topology is to be
distinguished from automatic generation of abstract topology. While
abstract topology is generated and updated automatically by
configuration as explained in Section 6.3.1, additional
supplementary topology may be obtained by the MDSC via path compute
request/reply mechanism. Starting with a black topology
advertisement from the PNCs, the MDSC may need additional
information beyond the level of black topology from the PNCs.
It is assumed that the black topology advertisement from PNCs would
give the MDSC each domain's the border node/link information. Under
this scenario, when the MDSC needs to allocate a new VN, the MDSC
can issue a number of Path Computation requests as described in
[ACTN-YANG] to different PNCs with constraints matching the VN
request. An example is provided in Figure 4, where the MDSC is
requesting to setup a P2P VN between AP1 and AP2. The MDSC can use
two different inter-domain links to get from Domain X to Domain Y,
namely the one between ASBRX.1 and ASBRY.1 and the one between
ASBRX.2 and ASBRY.2, but in order to choose the best end to end path
it needs to know what domain X and Y can offer in term of
connectivity and constraints between the PE nodes and the ASBR
nodes.
------- -------
( ) ( )
- ASBRX.1------- ASBRY.1 -
(+---+ ) ( +---+)
-+---( |PE1| Dom.X ) ( Dom.Y |PE2| )---+-
| (+---+ ) ( +---+) |
AP1 - ASBRX.2------- ASBRY.2 - AP2
( ) ( )
------- -------
Figure 4: A multi-domain networks example
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 28]
Internet-Draft ACTN Framework July 2017
A path computation request will be issued to PNC.X asking for
potential connectivity between PE1 and ASBRX.1 and between PE1 and
ASBRX.2 with related objective functions and TE metric constraints.
A similar request will be issued to PNC.Y and the results merged
together at the MDSC to be able to compute the optimal end-to-end
path including the inter domain links.
The info related to the potential connectivity may be cached by the
MDSC for subsequent path computation processes or discarded, but in
this case the PNCs are not requested to keep the grey topology
updated.
6.4. Abstraction Configuration Consideration
This section provides a set of abstraction configuration
considerations.
It is expected that the abstraction level be configured between the
CNC and the MDSC (i.e., the CMI) depending on the capability of the
CNC. This negotiated level of abstraction on the CMI may also impact
the way the MDSC and the PNCs configure and encode the abstracted
topology. For example, if the CNC is capable of sophisticated
technology specific operation, then this would impact the level of
abstraction at the MDSC with the PNCs. On the other hand, if the CNC
asks for a generic topology abstraction, then the level of
abstraction at the MDSC with the PNCs can be less technology
specific than the former case.
The subsequent sections provide a list of possible abstraction
levels for various technology domain networks.
6.4.1. Packet Networks
- For grey abstraction, the type of abstraction and its parameters
can be defined and configured.
o Abstraction Level 1: TE-tunnel abstraction for all (S-D)
border pairs with:
. Maximum B/W available per Priority Level
. Minimum Latency
6.4.2. OTN Networks
For OTN networks, max bandwidth available may be per ODU 0/1/2/3
switching level or aggregated across all ODU switching levels (i.e.,
ODUj/k). Clearly, there is a trade-off between these two abstraction
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 29]
Internet-Draft ACTN Framework July 2017
methods. Some OTN switches can switch any level of ODUs and in such
case there is no need for ODU level abstraction.
- For grey abstraction, the type of abstraction and its parameters
can be defined and configured.
o Abstraction Level 1: Per ODU Switching level (i.e., ODU type
and number) TE-tunnel abstraction for all (S-D) border pairs
with:
. Maximum B/W available per Priority Level
. Minimum Latency
o Abstraction Level 2: Aggregated TE-tunnel abstraction for all
(S-D) border pairs with:
. Maximum B/W available per Priority Level
. Minimum Latency
6.4.3. WSON Networks
For WSON networks, max bandwidth available may be per
lambda/frequency level (OCh) or aggregated across all
lambda/frequency level. Per OCh level abstraction gives more
detailed data to the MDSC at the expense of more information
processing. Either OCh-level or aggregated level abstraction should
factor in the RWA constraint (i.e., wavelength continuity) at the
PNC level. This means the PNC should have this capability and
advertise it as such.
For grey abstraction, the type of abstraction and its parameters can
be defined and configured as follows:
o Abstraction Level 1: Per Lambda/Frequency level TE-tunnel
abstraction for all (S-D) border pairs with:
. Maximum B/W available per Priority Level
. Minimum Latency
o Abstraction Level 2: Aggregated TE-tunnel abstraction for all
(S-D) border pairs with:
. Maximum B/W available per Priority Level
6.5. Topology Abstraction Granularity Level example
This section illustrates how topology abstraction operates in
different level of granularity over a hierarchy of MDSCs which is
shown in Figure 8 below.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 30]
Internet-Draft ACTN Framework July 2017
+-----+
| CNC | CNC wants to create a VN
+-----+ between CE A and CE B
|
|
+-----------------------+
| MDSC-H |
+-----------------------+
/ \
/ \
+--------+ +--------+
| MDSC-L1| | MDSC-L2|
+--------+ +--------+
/ \ / \
/ \ / \
+-----+ +-----+ +-----+ +-----+
CE A o----|PNC 1| |PNC 2| |PNC 3| |PNC 4|----o CE B
+-----+ +-----+ +-----+ +-----+
Topology operated by MDSC-H
--o=o=o=o--
Topology operated by MDSC-L1 Topology operated by MDSC-L2
_ _ _ _
( ) ( ) ( ) ( )
( ) ( ) ( ) ( )
--(o---o)==(o---o)== ==(o---o)==(o---o)--
( ) ( ) ( ) ( )
(_) (_) (_) (_)
Actual Topology
___ ___ ___ ___
( ) ( ) ( ) ( )
( o ) ( o ) ( o--o) ( o )
( / \ ) ( |\ ) ( | | ) ( / \ )
----(o-o---o-o)==(o-o-o-o-o)==(o--o--o-o)==(o-o-o-o-o)----
( \ / ) ( | |/ ) ( | | ) ( \ / )
( o ) (o-o ) ( o--o) ( o )
(___) (___) (___) (___)
Domain 1 Domain 2 Domain 3 Domain 4
Where o is a node and -- is a link and === a border link
Figure 8: Illustration of topology abstraction granularity levels
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 31]
Internet-Draft ACTN Framework July 2017
In the example depicted in Figure 8, there are four domains under
control of the respective PNCs, namely, PNC 1, PNC 2, PNC3 and PNC4.
Assume that MDSC L-1 is controlling PNC 1 and PNC 2 while MDSC L-2
is controlling PNC 3 and PNC 4. Let us assume that each of the PNCs
provides a grey topology abstraction in which to present only border
nodes and links within and outside the domain. The abstract topology
MDSC-L1 would operate is basically a combination of the two
topologies the PNCs (PNC 1 and PNC 2) provide. Likewise, the
abstract topology MDSC-L2 would operate is shown in Figure 8. Both
MDSC-L1 and MDSC-L2 provide a black topology abstraction in which
each PNC domain is presented as one virtual node to its top level
MDSC-H. Then the MDSC-H combines these two topologies updated by
MDSC-L1 and MDSC-L2 to create the abstraction topology to which it
operates. MDSC-H sees the whole four domain networks as four virtual
nodes connected via virtual links. The top level MDSC may operate on
a higher level of abstraction (i.e., less granular level) than the
lower level MSDCs.
7. Access Points and Virtual Network Access Points
In order not to share unwanted topological information between the
customer domain and provider domain, a new entity is defined which
is referred to as the Access Point (AP). See the definition of AP in
Section 1.1.
A customer node will use APs as the end points for the request of
VNS as shown in Figure 9.
-------------
( )
- -
+---+ X ( ) Z +---+
|CE1|---+----( )---+---|CE2|
+---+ | ( ) | +---+
AP1 - - AP2
( )
-------------
Figure 9: APs definition customer view
Let's take as an example a scenario shown in Figure 7. CE1 is
connected to the network via a 10Gb link and CE2 via a 40Gb link.
Before the creation of any VN between AP1 and AP2 the customer view
can be summarized as shown in Table 1:
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 32]
Internet-Draft ACTN Framework July 2017
+----------+------------------------+
|End Point | Access Link Bandwidth |
+-----+----------+----------+-------------+
|AP id| CE,port | MaxResBw | AvailableBw |
+-----+----------+----------+-------------+
| AP1 |CE1,portX | 10Gb | 10Gb |
+-----+----------+----------+-------------+
| AP2 |CE2,portZ | 40Gb | 40Gb |
+-----+----------+----------+-------------+
Table 1: AP - customer view
On the other hand, what the provider sees is shown in Figure 10.
------- -------
( ) ( )
- - - -
W (+---+ ) ( +---+) Y
-+---( |PE1| Dom.X )----( Dom.Y |PE2| )---+-
| (+---+ ) ( +---+) |
AP1 - - - - AP2
( ) ( )
------- -------
Figure 10: Provider view of the AP
Which results in a summarization as shown in Table 2.
+----------+------------------------+
|End Point | Access Link Bandwidth |
+-----+----------+----------+-------------+
|AP id| PE,port | MaxResBw | AvailableBw |
+-----+----------+----------+-------------+
| AP1 |PE1,portW | 10Gb | 10Gb |
+-----+----------+----------+-------------+
| AP2 |PE2,portY | 40Gb | 40Gb |
+-----+----------+----------+-------------+
Table 2: AP - provider view
A Virtual Network Access Point (VNAP) needs to be defined as binding
between the AP that is linked to a VN and that is used to allow for
different VNs to start from the same AP. It also allows for traffic
engineering on the access and/or inter-domain links (e.g., keeping
track of bandwidth allocation). A different VNAP is created on an AP
for each VN.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 33]
Internet-Draft ACTN Framework July 2017
In the simple scenario depicted above we suppose we want to create
two virtual networks. The first with VN identifier 9 between AP1 and
AP2 with bandwidth of 1Gbps, while the second with VN id 5, again
between AP1 and AP2 and with bandwidth 2Gbps.
The provider view would evolve as shown in Table 3.
+----------+------------------------+
|End Point | Access Link/VNAP Bw |
+---------+----------+----------+-------------+
|AP/VNAPid| PE,port | MaxResBw | AvailableBw |
+---------+----------+----------+-------------+
|AP1 |PE1,portW | 10Gbps | 7Gbps |
| -VNAP1.9| | 1Gbps | N.A. |
| -VNAP1.5| | 2Gbps | N.A |
+---------+----------+----------+-------------+
|AP2 |PE2,portY | 40Gbps | 37Gbps |
| -VNAP2.9| | 1Gbps | N.A. |
| -VNAP2.5| | 2Gbps | N.A |
+---------+----------+----------+-------------+
Table 3: AP and VNAP - provider view after VNS creation
7.1. Dual homing scenario
Often there is a dual homing relationship between a CE and a pair of
PEs. This case needs to be supported by the definition of VN, APs
and VNAPs. Suppose CE1 connected to two different PEs in the
operator domain via AP1 and AP2 and that the customer needs 5Gbps of
bandwidth between CE1 and CE2. This is shown in Figure 11.
____________
AP1 ( ) AP3
-------(PE1) (PE3)-------
W / ( ) \X
+---+/ ( ) \+---+
|CE1| ( ) |CE2|
+---+\ ( ) /+---+
Y \ ( ) /Z
-------(PE2) (PE4)-------
AP2 (____________)
Figure 11: Dual homing scenario
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 34]
Internet-Draft ACTN Framework July 2017
In this case, the customer will request for a VN between AP1, AP2
and AP3 specifying a dual homing relationship between AP1 and AP2.
As a consequence no traffic will flow between AP1 and AP2. The dual
homing relationship would then be mapped against the VNAPs (since
other independent VNs might have AP1 and AP2 as end points).
The customer view would be shown in Table 4.
+----------+------------------------+
|End Point | Access Link/VNAP Bw |
+---------+----------+----------+-------------+-----------+
|AP/VNAPid| CE,port | MaxResBw | AvailableBw |Dual Homing|
+---------+----------+----------+-------------+-----------+
|AP1 |CE1,portW | 10Gbps | 5Gbps | |
| -VNAP1.9| | 5Gbps | N.A. | VNAP2.9 |
+---------+----------+----------+-------------+-----------+
|AP2 |CE1,portY | 40Gbps | 35Gbps | |
| -VNAP2.9| | 5Gbps | N.A. | VNAP1.9 |
+---------+----------+----------+-------------+-----------+
|AP3 |CE2,portX | 40Gbps | 35Gbps | |
| -VNAP3.9| | 5Gbps | N.A. | NONE |
+---------+----------+----------+-------------+-----------+
Table 4: Dual homing - customer view after VN creation
8. Advanced ACTN Application: Multi-Destination Service
A further advanced application of ACTN is in the case of Data Center
selection, where the customer requires the Data Center selection to
be based on the network status; this is referred to as Multi-
Destination in [ACTN-REQ]. In terms of ACTN, a CNC could request a
connectivity service (virtual network) between a set of source Aps
and destination APs and leave it up to the network (MDSC) to decide
which source and destination access points to be used to set up the
connectivity service (virtual network). The candidate list of source
and destination APs is decided by a CNC (or an entity outside of
ACTN) based on certain factors which are outside the scope of ACTN.
Based on the AP selection as determined and returned by the network
(MDSC), the CNC (or an entity outside of ACTN) should further take
care of any subsequent actions such as orchestration or service
setup requirements. These further actions are outside the scope of
ACTN.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 35]
Internet-Draft ACTN Framework July 2017
Consider a case as shown in Figure 12, where three data centers are
available, but the customer requires the data center selection to be
based on the network status and the connectivity service setup
between the AP1 (CE1) and one of the destination APs (AP2 (DC-A),
AP3 (DC-B), and AP4 (DC-C)). The MDSC (in coordination with PNCs)
would select the best destination AP based on the constraints,
optimization criteria, policies, etc., and setup the connectivity
service (virtual network).
------- -------
( ) ( )
- - - -
+---+ ( ) ( ) +----+
|CE1|---+----( Domain X )----( Domain Y )---+---|DC-A|
+---+ | ( ) ( ) | +----+
AP1 - - - - AP2
( ) ( )
---+--- ---+---
AP3 | AP4 |
+----+ +----+
|DC-B| |DC-C|
+----+ +----+
Figure 12: End point selection based on network status
8.1. Pre-Planned End Point Migration
Further in case of Data Center selection, customer could request for
a backup DC to be selected, such that in case of failure, another DC
site could provide hot stand-by protection. As shown in Figure 13
DC-C is selected as a backup for DC-A. Thus, the VN should be setup
by the MDSC to include primary connectivity between AP1 (CE1) and
AP2 (DC-A) as well as protection connectivity between AP1 (CE1) and
AP4 (DC-C).
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 36]
Internet-Draft ACTN Framework July 2017
------- -------
( ) ( )
- - __ - -
+---+ ( ) ( ) +----+
|CE1|---+----( Domain X )----( Domain Y )---+---|DC-A|
+---+ | ( ) ( ) | +----+
AP1 - - - - AP2 |
( ) ( ) |
---+--- ---+--- |
AP3 | AP4 | HOT STANDBY
+----+ +----+ |
|DC-D| |DC-C|<-------------
+----+ +----+
Figure 13: Pre-planned end point migration
8.2. On the Fly End Point Migration
Compared to pre-planned end point migration, on the fly end point
selection is dynamic in that the migration is not pre-planned but
decided based on network condition. Under this scenario, the MDSC
would monitor the network (based on the VN SLA) and notify the CNC
in case where some other destination AP would be a better choice
based on the network parameters. The CNC should instruct the MDSC
when it is suitable to update the VN with the new AP if it is
required.
9. Advanced Topic
This section describes how ACTN architecture supports some
deployment scenarios. See Appendix A for details on MDSC and PNC
functions integrated in Service/Network Orchestrator and Appendix B
for IP + Optical with L3VPN service.
10. Manageability Considerations
The objective of ACTN is to manage traffic engineered resources, and
provide a set of mechanism to allow clients to request virtual
connectivity across server network resources. ACTN will support
multiple clients each with its own view of and control of the server
network, the network operator will need to partition (or "slice")
their network resources, and manage them resources accordingly.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 37]
Internet-Draft ACTN Framework July 2017
The ACTN platform will, itself, need to support the request,
response, and reservations of client and network layer connectivity.
It will also need to provide performance monitoring and control of
traffic engineered resources. The management requirements may be
categorized as follows:
. Management of external ACTN protocols
. Management of internal ACTN protocols
. Management and monitoring of ACTN components
. Configuration of policy to be applied across the ACTN system
10.1. Policy
It is expected that a policy will be an important aspect of ACTN
control and management. Typically, policies are used via the
components and interfaces, during deployment of the service, to
ensure that the service is compliant with agreed policy factors
(often described in Service Level Agreements - SLAs), these include,
but are not limited to: connectivity, bandwidth, geographical
transit, technology selection, security, resilience, and economic
cost.
Depending on the deployment the ACTN deployment architecture, some
policies may have local or global significance. That is, certain
policies may be ACTN component specific in scope, while others may
have broader scope and interact with multiple ACTN components. Two
examples are provided below:
. A local policy might limit the number, type, size, and
scheduling of virtual network services a customer may request
via its CNC. This type of policy would be implemented locally on
the MDSC.
. A global policy might constrain certain customer types (or
specific customer applications) to only use certain MDSCs, and
be restricted to physical network types managed by the PNCs. A
global policy agent would govern these types of policies.
This objective of this section is to discuss the applicability of
ACTN policy: requirements, components, interfaces, and examples.
This section provides an analysis and does not mandate a specific
method for enforcing policy, or the type of policy agent that would
be responsible for propagating policies across the ACTN components.
It does highlight examples of how policy may be applied in the
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 38]
Internet-Draft ACTN Framework July 2017
context of ACTN, but it is expected further discussion in an
applicability or solution specific document, will be required.
10.2. Policy applied to the Customer Network Controller
A virtual network service for a customer application will be
requested from the CNC. It will reflect the application requirements
and specific service policy needs, including bandwidth, traffic type
and survivability. Furthermore, application access and type of
virtual network service requested by the CNC, will be need adhere to
specific access control policies.
10.3. Policy applied to the Multi Domain Service Coordinator
A key objective of the MDSC is to help the customer express the
application connectivity request via its CNC as set of desired
business needs, therefore policy will play an important role.
Once authorized, the virtual network service will be instantiated
via the CNC-MDSC Interface (CMI), it will reflect the customer
application and connectivity requirements, and specific service
transport needs. The CNC and the MDSC components will have agreed
connectivity end-points, use of these end-points should be defined
as a policy expression when setting up or augmenting virtual network
services. Ensuring that permissible end-points are defined for CNCs
and applications will require the MDSC to maintain a registry of
permissible connection points for CNCs and application types.
It may also be necessary for the MDSC to resolve policy conflicts,
or at least flag any issues to administrator of the MDSC itself.
Conflicts may occur when virtual network service optimization
criterion are in competition. For example, to meet objectives for
service reachability a request may require an interconnection point
between multiple physical networks; however, this might break a
confidentially policy requirement of specific type of end-to-end
service. This type of situation may be resolved using hard and soft
policy constraints.
10.4. Policy applied to the Physical Network Controller
The PNC is responsible for configuring the network elements,
monitoring physical network resources, and exposing connectivity
(direct or abstracted) to the MDSC. It is therefore expected that
policy will dictate what connectivity information will be exported
between the PNC, via the MDSC-PNC Interface (MPI), and MDSC.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 39]
Internet-Draft ACTN Framework July 2017
Policy interactions may arise when a PNC determines that it cannot
compute a requested path from the MDSC, or notices that (per a
locally configured policy) the network is low on resources (for
example, the capacity on key links become exhausted). In either
case, the PNC will be required to notify the MDSC, which may (again
per policy) act to construct a virtual network service across
another physical network topology.
Furthermore, additional forms of policy-based resource management
will be required to provide virtual network service performance,
security and resilience guarantees. This will likely be implemented
via a local policy agent and subsequent protocol methods.
11. Security Considerations
The ACTN framework described in this document defines key components
and interfaces for managed traffic engineered networks. Securing the
request and control of resources, confidentially of the information,
and availability of function, should all be critical security
considerations when deploying and operating ACTN platforms.
Several distributed ACTN functional components are required, and as
a rule implementations should consider encrypting data that flow
between components, especially when they are implemented at remote
nodes, regardless if these are external or internal network
interfaces.
The ACTN security discussion is further split into two specific
categories described in the following sub-sections:
. Interface between the Customer Network Controller and Multi
Domain Service Coordinator (MDSC), CNC-MDSC Interface (CMI)
. Interface between the Multi Domain Service Coordinator and
Physical Network Controller (PNC), MDSC-PNC Interface (MPI)
From a security and reliability perspective, ACTN may encounter many
risks such as malicious attack and rogue elements attempting to
connect to various ACTN components. Furthermore, some ACTN
components represent a single point of failure and threat vector,
and must also manage policy conflicts, and eavesdropping of
communication between different ACTN components.
The conclusion is that all protocols used to realize the ACTN
framework should have rich security features, and customer,
application and network data should be stored in encrypted data
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 40]
Internet-Draft ACTN Framework July 2017
stores. Additional security risks may still exist. Therefore,
discussion and applicability of specific security functions and
protocols will be better described in documents that are use case
and environment specific.
11.1. Interface between the Customer Network Controller and Multi
Domain Service Coordinator (MDSC), CNC-MDSC Interface (CMI)
The role of the MDSC is to detach the network and service control
from underlying technology to help the customer express the network
as desired by business needs. It should be noted that data stored by
the MDSC will reveal details of the virtual network services, and
which CNC and application is consuming the resource. The data stored
must therefore be considered as a candidate for encryption.
CNC Access rights to an MDSC must be managed. MDSC resources must be
properly allocated, and methods to prevent policy conflicts,
resource wastage and denial of service attacks on the MDSC by rogue
CNCs, should also be considered.
A CNC-MDSC protocol interface will likely be an external protocol
interface. Again, suitable authentication and authorization of each
CNC connecting to the MDSC will be required, especially, as these
are likely to be implemented by different organizations and on
separate functional nodes. Use of the AAA-based mechanisms would
also provide role-based authorization methods, so that only
authorized CNC's may access the different functions of the MDSC.
11.2. Interface between the Multi Domain Service Coordinator and
Physical Network Controller (PNC), MDSC-PNC Interface (MPI)
The function of the Physical Network Controller (PNC) is to
configure network elements, provide performance and monitoring
functions of the physical elements, and export physical topology
(full, partial, or abstracted) to the MDSC.
Where the MDSC must interact with multiple (distributed) PNCs, a
PKI-based mechanism is suggested, such as building a TLS or HTTPS
connection between the MDSC and PNCs, to ensure trust between the
physical network layer control components and the MDSC.
Which MDSC the PNC exports topology information to, and the level of
detail (full or abstracted) should also be authenticated and
specific access restrictions and topology views, should be
configurable and/or policy-based.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 41]
Internet-Draft ACTN Framework July 2017
12. References
12.1. Informative References
[RFC2702] Awduche, D., et. al., "Requirements for Traffic
Engineering Over MPLS", RFC 2702, September 1999.
[RFC4026] L. Andersson, T. Madsen, "Provider Provisioned Virtual
Private Network (VPN) Terminology", RFC 4026, March 2005.
[RFC4208] G. Swallow, J. Drake, H.Ishimatsu, Y. Rekhter,
"Generalized Multiprotocol Label Switching (GMPLS) User-
Network Interface (UNI): Resource ReserVation Protocol-
Traffic Engineering (RSVP-TE) Support for the Overlay
Model", RFC 4208, October 2005.
[RFC4655] Farrel, A., Vasseur, J.-P., and J. Ash, "A Path
Computation Element (PCE)-Based Architecture", IETF RFC
4655, August 2006.
[RFC5654] Niven-Jenkins, B. (Ed.), D. Brungard (Ed.), and M. Betts
(Ed.), "Requirements of an MPLS Transport Profile", RFC
5654, September 2009.
[RFC7149] Boucadair, M. and Jacquenet, C., "Software-Defined
Networking: A Perspective from within a Service Provider
Environment", RFC 7149, March 2014.
[RFC7926] A. Farrel (Ed.), "Problem Statement and Architecture for
Information Exchange between Interconnected Traffic-
Engineered Networks", RFC 7926, July 2016.
[GMPLS] Manning, E., et al., "Generalized Multi-Protocol Label
Switching (GMPLS) Architecture", RFC 3945, October 2004.
[ONF-ARCH] Open Networking Foundation, "SDN architecture", Issue
1.1, ONF TR-521, June 2016.
[RFC7491] King, D., and Farrel, A., "A PCE-based Architecture for
Application-based Network Operations", RFC 7491, March
2015.
[Transport NBI] Busi, I., et al., "Transport North Bound Interface
Use Cases", draft-tnbidt-ccamp-transport-nbi-use-cases,
work in progress.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 42]
Internet-Draft ACTN Framework July 2017
[ACTN-Abstraction] Y. Lee, et al., "Abstraction and Control of TE
Networks (ACTN) Abstraction Methods", draft-lee-teas-actn-
abstraction, work in progress.
13. Contributors
Adrian Farrel
Old Dog Consulting
Email: adrian@olddog.co.uk
Italo Busi
Huawei
Email: Italo.Busi@huawei.com
Khuzema Pithewan
Infinera
Email: kpithewan@infinera.com
Michael Scharf
Nokia
Email: michael.scharf@nokia.com
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 43]
Internet-Draft ACTN Framework July 2017
Authors' Addresses
Daniele Ceccarelli (Editor)
Ericsson
Torshamnsgatan,48
Stockholm, Sweden
Email: daniele.ceccarelli@ericsson.com
Young Lee (Editor)
Huawei Technologies
5340 Legacy Drive
Plano, TX 75023, USA
Phone: (469)277-5838
Email: leeyoung@huawei.com
Luyuan Fang
Microsoft
Email: luyuanf@gmail.com
Diego Lopez
Telefonica I+D
Don Ramon de la Cruz, 82
28006 Madrid, Spain
Email: diego@tid.es
Sergio Belotti
Alcatel Lucent
Via Trento, 30
Vimercate, Italy
Email: sergio.belotti@nokia.com
Daniel King
Lancaster University
Email: d.king@lancaster.ac.uk
Dhruv Dhody
Huawei Technologies
Divyashree Techno Park, Whitefield
Bangalore, Karnataka 560066
India
Email: dhruv.ietf@gmail.com
Gert Grammel
Juniper Networks
Email: ggrammel@juniper.net
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 44]
Internet-Draft ACTN Framework July 2017
APPENDIX A - Example of MDSC and PNC functions integrated in
Service/Network Orchestrator
This section provides an example of a possible deployment scenario,
in which Service/Network Orchestrator can include a number of
functionalities, among which, in the example below, PNC
functionalities for domain 2 and MDSC functionalities to coordinate
the PNC1 functionalities (hosted in a separate domain controller)
and PNC2 functionalities (co-hosted in the network orchestrator).
Customer
+-------------------------------+
| +-----+ |
| | CNC | |
| +-----+ |
+-------|-----------------------+
|-CMI
Service/Network |
Orchestrator |
+-------|------------------------+
| +------+ MPI +------+ |
| | MDSC |----|--> | PNC2 | |
| +------+ +------+ |
+-------|------------------|-----+
|-MPI |
Domain Controller | |
+-------|-----+ |
| +-----+ | |
| |PNC1 | | |
| +-----+ | |
+-------|-----+ |
v v
------- -------
( ) ( )
- - - -
( ) ( )
( Domain 1 )----( Domain 2 )
( ) ( )
- - - -
( ) ( )
------- -------
APPENDIX B - Example of IP + Optical network with L3VPN service
This section provides a more complex deployment scenario in which
ACTN hierarchy is deployed to control a multi-layer network via an
IP/MPLS PNC and an Optical PNC. The scenario is further enhanced by
the introduction of an upper layer service configuration (e.g.
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 45]
Internet-Draft ACTN Framework July 2017
L3VPN). The provisioning of the L3VPN service is outside ACTN scope
but it is worth showing how the two parts are integrated for the end
to end service fulfilment. An example of service configuration
function in the Service/Network Orchestrator is discussed in [I-
D.dhjain-bess-bgp-l3vpn-yang].
Customer
+-------------------------------+
| +-----+ |
| | CNC | |
| +-----+ |
+-------|--------+--------------+
|-CMI | Customer Service Model
| | (non-ACTN interface)
Service/Network | |
Orchestrator | |
+-------|--------|--------------------------+
| | +-------------------------+ |
| | |Service Mapping Function | |
| | +-------------------------+ |
| | | | |
| +------+ | +---------------+ |
| | MDSC |--- |Service Config.| |
| +------+ +---------------+ |
+------|------------------|-----------------+
MPI-| +------------+ (non-ACTN Interf.)
| /
+-----------/------------+
IP/MPLS | / |
Domain | / | Optical Domain
Controller | / | Controller
+--------|-------/----+ +---|--------------+
| +-----+ +-----+ | | +-----+ |
| |PNC1 | |Serv.| | | |PNC2 | |
| +-----+ +-----+ | | +-----+ |
+---------------------+ +------------------+
| |
v |
+---------------------------------+ |
/ IP/MPLS Network \ |
+-------------------------------------+ |
V
+--------------------------------------+
/ Optical Network \
+------------------------------------------+
Ceccarelli, Lee, et al. Expires January 19, 2018 [Page 46]