Deployment Guide

12 ScaleIO/VxFlex OS IP Fabric Best Practice and Deployment Guide with OS10EE | version 1.0
2.3 Routing protocol selection for Leaf-Spine
Typical routing protocols used when designing a Leaf-Spine network.
• Border Gateway Protocol (BGP) - External (EBGP) or Internal (IBGP)
• Open Shortest Path First (OSPF)
This guide includes examples for both OSPF and EBGP configurations.
Table 1 lists items to consider when choosing between OSPF and BGP protocols.
OSPF and BGP table
OSPF
BGP
OSPF is the choice of protocol for networks under
same administration or internal networks. OSPF is
an internal gateway protocol.
BGP is the choice of protocol for networks under
different administration or different ISPs. BGP is an
external gateway protocol.
OSPF is a link state protocol. OSPF chooses the
best path among all available paths and requires
no additional configurations as the protocol
dynamically reacts to any changes in paths.
BGP is a path vector routing protocol. It makes routing
decisions based on paths, network policies or rule-
sets configured by a network administrator and is
involved in making core routing decisions.
An OSPF network can be divided into sub-
domains called areas. An area is a logical
collection of OSPF networks, routers and links
that have the same area identification. A router
within an area must maintain a topological
database for the area to which it belongs. The
router does not have detailed information about
network topology outside of its area, thereby
reducing the size of its database.
When BGP runs between two peers in the same
autonomous system (AS), it is referred to as Internal
BGP (iBGP or Interior Border Gateway Protocol).
When it runs between different autonomous systems,
it is called External BGP (eBGP or Exterior Border
Gateway Protocol).
OSPF works within a single autonomous system.
BGP works with multiple autonomous systems.
Route filtering is not possible.
Route filtering is possible in BGP through Network
Layer Reachability, AS_Path and Community
attributes.
OSPF cannot scale easily in large networks.
BGP can scale in very large networks.
2.3.1 BGP
BGP provides scalability whether configured as EBGP or IBGP. See section 3.2.3 for an EBGP routing
configuration example on a Leaf-Spine network. The nature of a Leaf-Spine network is based on the use of
ECMP. EBGP and IBGP handle ECMP differently. By default, EBGP supports ECMP without any
adjustments. However, IBGP requires a BGP route reflector and the use of the AddPath feature to support
ECMP.