1. Network Hierarchy and Multilayer Survivability
<draft-team-restore-hierarchy-00.txt>
TEWG Requirements Design Team
J. Boyle, M. Carlzon, R. Coltun, T. Griffin
E. Kern, W.S. Lai, D. McDysan, T. Reddington
Objective: determine the current and near term requirements for survivability and hierarchy in multi-vendor service provider environments
Feedback from working group solicited
Definitions
Requirements

2. Hierarchy Horizontally oriented hierarchy
between two areas or administrative subdivisions within the same network layer
the abstraction necessary to allow a network at one network layer, for instance a packet network, to grow. Examples of horizontal hierarchy include BGP and multi-area OSPF.
Vertically oriented hierarchy
between two network layers
the abstraction, or reduction in information, which would be of benefit when communicating information across network layers, as in propagating information between optical transport layer and packet/router layer networks.

3. Survivability
the capability of a network to maintain service continuity in the presence of faults within the network.
Protection
based on predetermined failure recovery: as the working entity is established, resources are reserved for the protection entity
Restoration
mainly relies on the use of spare capacity in the network
Trade-offs
protection: usually offers fast recovery from failure with enhanced availability
restoration: usually achieves better resource utilization.

4. Survivability Requirements
need for common, interoperable survivability approaches in packet and non-packet networks
suggested survivability mechanisms
path-based
local repairs in proximity to the network fault
timing bounds for service restoration
voice call cutoff (140 msec to 2 sec)
protocol timer requirements in premium data services
mission critical applications
coordination among layers
currently based on use of nested timers
issue: may limit the recovery time for fast MPLS restoration if SDH/SONET protection switching must be completed first

5. Hierarchy Requirements
Horizontally oriented hierarchy (intra-domain)
mechanisms for edge-to-edge signaling, in the context of layer-2 and layer-3 VPN services where SLAs would appear to necessitate signaling from the edges into the core of a network
issues
potential current protocols limitations in networks which are hierarchical (e.g. multi-area OSPF)
scalability concerns of potentially O(N^2) connection growth in larger networks
Vertical oriented hierarchy
instead of direct exchange of signaling and routing between vertical layers, some looser form of coordination and communication is a nearer term need

6. Survivability Requirements in Horizontal Hierarchy
Protected end-to-end connection
a concatenated set of connections that are each protected within their area
Connection routing
a network element that participates on both sides of a boundary (e.g., OSPF ABR) - note that this is a common point of failure
route server
Inter-area signaling of survivability information
to enable a “least common denominator” survivability mechanism at the boundary
to convey the success or failure of the service restoration action; e.g., if a part of a "connection" is down on one side of a boundary, there is no need for the other side to recover from failures

7. Next Step Discussion New version of draft
incorporate comments received to clarify some of the text
refine definitions of terminology
include results in the discussion of this meeting
Discussion
future plan of the design team, target date for completion
any issues with current draft
kick off work in tewg and other working groups on more specific requirements drafts and protocol specifications drafts