Chapter 5: Designing a Network Topology
● Topology is defined as a branch of mathematics, the study of geometric properties and
spatial relations unaffected by the continuous change of shape or size of figures.
● Types; Bus topology, uses a common backbone to connect all devices, single cable
used, it is full duplex (both parties can talk at once).
● The receipts of a packet ignore it if not addressed to them, at the end of the bus there is
a terminator which absorbs any signal and removes the signal from the bus.
Advantages, usually easy to connect, requires less wiring and cheap.Disadvantages,
network shuts down if there is a break in a wire, hard to identify the problem if the
network shuts down. Uses CMSA/CD to prevent collisions.
● Star topology; each station is connected to common central node, has a central
connection called Hub node, could be a hub or switch. Devices connect to a hub using
UTP ethernet. Requires more cables, a failure in star will only take out 1 computer. If the
hub fails the entire network fails.
● Ring, consists of set of repeaters that are joined point to point in a closed loop. Linkes
are unidirectional, one way.
● Tree, hybrid topology mix of bus and and bus or bus and star. A large computer network
could be constructed but it relies on central hub, if the hub fails the entire network fails.
● Mesh, devices are connected with many redundant interconnections, between network
nodes. 2 types partial and full mesh, full mesh every node is connected to another node,
formula is n* (n-1)/2 , partial mesh not all are connected is cheaper and less redundant.
● Hierarchical network design model help you develop a topology in discrete layers. Each
layer can be focused on specific functions.
● Typical topology consists of core layer; high end routers and switches that are optimized
for availability and performance. Distribution layer of routers and switches that
implement policies and access layer that connects users via lower end switches and
wireless access points.
● Furball networks that grow unheeded without any plan in place, unstructured.
● Why hierarchical, because it reduces workload on a network, facilitates changes,
● Flat vs Hierarchical topologies, flat is adequate for small networks, there is no hierarchy.
Easy to design, easy to implement and maintain as long as its small. When it grows flat
is no longer desirable, troubleshooting becomes difficult. Flat WAN few sites connected
in a loop, to two other adjacents via point to point links.
● Flat LAN, segment the network into smaller bandwidths, all devices are in the same
background domains. Adequate for small lans.
● Mesh vs Hierarchical Mesh topologies uses a hub and spoke topology approach to
● 3 layer hierarchical model, each layer has a specific role. The core layer provides
optimal transport between sites. Distribution layer connects network services to the
access layer and implements policies regarding security, loading and routing. Access
layer consist of routers at the edge of the campus network. ● Core Layer; known as the high speed backbone of the internetwork. Critical for
interconnectivity, should be designed with redundant components. Provides transports
between sites, should have a limited size for components, Services like internet, DMZ ,
and RAS should be centralized.
● Distribution Layer; the demarcation point between the access and core layers of the
network. Many roles, including controlling access to resources for security reasons and
controlling network traffic. Hides details about access layer from the core.
● Access Layer; provides users on local segments with access to the internetwork, can
include routers, switches and bridges, shared- media hubs and wireless access points.
● Guidelines; first control the diameter of a hierarchical enterprise network topology, by
doing this provides low and predictable latency. Adding a chain is a common mistake is
when network administrator adds another branch.
● Also good to avoid backdoors, backdoor is a connection between devices in the same
layer. Can be an extra router, bridge or switch added to connect two networks.
● Finally design access layer first, distribution and then finally core layer last. Starting with
the access layer can help you accurately perform capacity planning.
● Redundant network topologies, primary goal is to meet availability requirements by
duplicating elements of the network. Eliminates single point of failure that could cripple
the network. Can be implemented at any point in a layer, mostly core and distribution.
● Backup paths, to maintain interconnectivity network designs have a backup path for
packets to travel when there are problems on the primary path. Consists of links and
routers and switches identical to the primary path. Two considerations to consider, how
much capacity, how quickly the network will begin to use the backup path. Must be
● Load sharing; a secondary goal is to improve performance by supporting load sharing
across parallel links. in WAN environments you do this by configuring channel
aggregation meaning a router can automatically bring up multiple channels as bandwidth
requirements increase. MPPP, multilink point to point protocol ensures that packets
arrive in sequence at the receiving router. To do this data is encapsulated with the point
to point protocol. Pinhole congestion is when a slow link becomes saturated, higher
capacity links cannot be filled.
● CISCO SAFE; reference architecture that network designers can use to simplify the
complexity of a large internetwork.
● Core; stitches together all other modules, high speed infrastructure that provides layer 2
and 3 transport. Data Center; hosts servers, applications and storage devices for internal
users and connected network infrastructure that these devices require. Campus;