Implementing Cisco Switched Networks Chapter 7 Review

This is my review and notes of Chapter 7 of “Implementing Cisco Switched Networks Foundation Learning Guide”.


Chapter 7: Preparing the Campus Infrastructure for Advanced Services

WAFFLE WAFFLE WAFFLE. This chapter felt like “oh man, we totally need another chapter!”. When this book was first written, a lot of the content of this chapter was probably in the CCNP Switch exam. It covers topics that I’ve understood are now hardly touched on at all, if at it. Not to say it doesn’t still have some important topics in it. I read this chapter mainly to make sure I’ve covered every possible thing, but a lot of it I didn’t make notes on.

This chapter really went on and on about things that are common sense, especially things you should take into consideration when deploying voice, or video.

It did have some good content, but man it felt like there was a lot of extra stuff that just doesn’t matter for the exam.

What’s probably going to be in the exam:
– Just know the very simple things like important commands such as AutoQoS
– Very basic understanding of QoS, e.g. what DSCP and CoS is
– Very basic understanding of PIM-SM, PIM-DM
– Very basic understanding of IGMP, and IGMP Snooping

What the chapter also covers:
– Preparing your network for Wireless, Voice, and Video
– There was A LOT on each of these

Initially chapter is basically a spiel on why converged networks with video, voice, along with decent wifi connectivity are good.

Then goes into an introduction to WiFi, and its’ application in the Campus Network. I don’t have too many notes on this, mainly because I believe the exam doesn’t heavily focus on wireless and voice any more, as these have become more separate certification tracks of their own.

One point to make about wireless is the two deployment methods:

– Standalone WLAN Solution (also referred to as autonomous)
– Controller-based WLAN Solution

Traffic handling in a Controller-Based Solution
– Data and control messages are encapsulated between the access point and the WLC using the Control and Provisioning of Wireless Access Pooints (CAPWAP) or the LIghtweight Access Point Protocol (LWAPP). Both methods are standards based, but only LWAPP was adopted by Cisco.
– Control packets are enapsulated within the LWAPP or CAPWAP and encrypted.
– ^Same with Data traffic. The traffic is then switched at the WLC.

WLC is typically deployed at the distribution layer.

Hybrid Remote Edge Access Points (HREAP)
– Useful for providing HA of controller-based wireless solutions in remote offices.
– The purpose of these access points is to still offer wireless client connectivity when their connection to the controller (WLC) is lost.

Planning for the Campus Network to Support Voice

Again a lot on planning, and a lot on general considerations when implementing Voice.

– Voice uses UDP
– Voice is very much time sensitive traffic and a lot of thought needs to go into it when implementing.

Planning for the Campus Network to Support Video

Again talks about how it’s time sensitive, but in this case also very bandwidth hungry. This section was still several pages long, but I didn’t really feel like I read anything worth writing down. It was all pretty general, but obvious things to consider.

From these topics we come to something we really need to understand, and that is QoS.

Understanding QoS

Quality of Service refers to the prioritisation of different types of traffic. An example of its’ use is time critical traffic such as Voice packets not being swamped and lost due to lower priority traffic such as P2P consuming all the links bandwidth.

QoS allows you to have:
– Control over resources – Classifying high priority and low priority traffic
– More efficient use of network resources – By using network analysers, you can see a view of the traffic flows within your network and dedicate a higher priority to network resources that require it.
– Tailored services – In an ISP setting, better QoS can be sold at a price to customers who need it.
– Coexistence of mission-critical applications – Correct QoS configuration can ensure mission-critical apps are given exactly what they need.

Network Stability Issues
– Delay (latency)
– Delay variation (Jitter)
– Packet loss

QoS Service Models

– Best-effort service
Basically no QoS is applied, and the switch simply does FIFO (First In, First Out), so packets are simply transmitted out as they arrive in a queue

– Integrated Services (IntServ)
AKA hard QoS.
Implies that traffic flows are reserved explicitly by all intermediate systems and resources

-Differentiated Services (DiffServ)
AKA soft QoS.
Class-based, which some classes of traffic receive preferential handling over other traffic classes
DiffServ categorizes traffic, and then sorts it into queues of various efficiencies.

Very handy Cisco feature that automates a lot of the QoS configuration

– Basically it does all the QoS config work for you, and therefore requires less of an understanding of QoS config, and enables more less experienced staff to deploy it.

AutoQoS covers:
– Application classification
– Policy generation
– Configuration
– Monitoring and reporting
– Consistency

Traffic Classification and Marking

– Layer 2 = CoS (Class of Service)

At Layer 2, 3 bits are available in 802.1Q frames for classification for up to eight distinct levels of service: 0 through to 7.

– Layer 3 = DSCP (Differentiated Services Code Point)

At Layer 3, QoS uses the six most significant ToS bits in the IP header for a DSCP field definition.
– Allows up to 64 distinct values (levls of service): 0 through to 63.
– Last 2 bits represent the Early Congestion Notification (ECN) bits.

– Distinguishes a frame or packet with a specific priority or predetermined criteria.
– For Cisco Catalyst switches, classification determines the internal DSCP value on frames. This is for internal QoS packet handling, including policing and scheduling as frames traverse the switch.

When planning QoS in the Campus Network, always apply QoS classification as close to the edge as possible, preferably in the access layer.

Trust Boundaries and Configurations

Covers how a frame is handled as it arrives in on the switch.

E.g. Switch receives a packet with a DSCP value of 46, the switch accepts the ingress QoS value, and internally also uses the same value.

Cisco switches support accepting DSCP, IP Precedence, or CoS values on ingress frames.

The following is the default mappings, for how each L2 and L3 classification is mapped between L2 and L3 respectively:




















IP Precedence


















Refers to changing the DSCP, CoS, or IP Precedence bits on ingress frames.
– Configurable on a per-interface basis via a policy-map.

Traffic Shaping and Policing

Shaping refers to metering traffic and buffering or delaying excessive traffic so that the traffic rate stays within a desired limit rate. Shaping therefore is not very good for delay-sensitive traffic flows such as voice, video, or storage, but is useful for typical, bursty TCP flows.

Policing differs in that it takes a specific action for out-of-profile traffic exceeding a specified limit.
Policing does not delay or buffer traffic.
Actions for exceeding traffic by default is block, but other options are available such as permissible, trusting, and marking.

Congestion Management

Covers several queuing mechanisms:
– FIFO queuing → switch places all egress frames into the same queue, regardless of classification. Packets are sent out in the same order that they are received.
– Weighted round robin (WRR) queueing → Popular and simple method of differentiating service among traffic classes. With WRR, the switch uses a configured weight value for each egress queue. This weight value determines the proportion of bandwidth of each queue.
– Priority queuing → Switch will process any traffic in the highest of queues before lower queues. Since this method can result in queue starvation in the nonpriority queues, the remaining queues are subject to the WRR queueing to avoid this issue.
– Custom queuing → Strictly for WAN.

Congestion Avoidance

Monitor network traffic loads in an effort to anticipate and avoid congestion at common network bottleneck points.

Then goes into some methods of collision avoidance such as Tail Drop, and Weighted Random Early Detection.

Introduction to IP Multicast in the Campus Network

Initially goes over some basic stuff about multicast, including application and different with unicast and broadcast.

Covers the different types of Multicast addresses:
– Reserved link local addresses – –
– Globally scoped addresses – –
– Source-specific multicast addresses to – PIM-SSM reserved
– GLOP addresses to
– Limited-scope addresses to

Multicast MAC Address Structure

Multicast IP addresses get mapped to L2 MAC addresses. Due to how the structuring of this works, each Multicast MAC address maps to 32 potential IP addresses. In other words, each multicast MAC address represents a possible 32 distinct IP multicast addresses.

Reverse Path Forwarding
Is a mechanism that performs an incoming interface check to determine whether to forward or drop an incoming multicast frame.

When a multicast router receives a multicast packet, it determines which direction is the upstream direction (toward the source) and which one is the downstream direction (toward the receiver). A router forwards multicast packets only if the packet is received on the correct upstream interface determined by the RPF process.

Multicast Forwarding Tree
– Source Trees
– Shared Trees

– Both source trees and shared trees avoid multicast traffic loops. Routing devices replicate the multicast packets only where the tree tranches.

Source Trees
– Maintains path information for each multicast source

Shared Trees
– Have the advantage of requiring the minimum amount of state information in each router.
– This lowers the overall memory requirements and complexity for a network that allows only shared trees.
– The disadvantage of shared trees is that, under certain circumstances, the paths between the source and receivers might not be the optimal path, which can introduce additional latency in packet delivery.
Shared trees might overuse some links and leave others unused. In comparison, source trees usually distribute traffic across a set of links.

IP Multicast Protocols

Similar to IP unicast, where it can its’ own routing, management, and L2 protocols.

This section covers:
– PIM (Protocol Independent Multicast) – L3
– IGMP (Internet Group Management Protocol) – L2

– A multicast routing protocol.
– Uses the current unicast routing protocol to forward IP multicast traffic to other PIM neighbors.

Has several different versions:
– PIM Dense Mode
– PIM Sparse Mode
– PIM Sparse-Dense Mode
– PIM bidirectional

Sparse-dense mode is most common in large enterprise networks.

PIM Dense Mode
– Relies on periodic flooding of the network with multicast traffic to set up and maintain the distribution tree.
– Generally considered obsolete and no longer deployed in campus networks.
– PIM-DM floods the multimedia packet to all routers in the network and then prunes routers that do not service members of that particular multicast group.

PIM Sparse Mode
– Based on the assumption that the multicast group members are sparsely distributed throughout the network and that bandwidth is limited.
– Important to note that PIM-SM does not imply that the group has few members, just that they are widely dispersed.

-PIM-SM differs from dense mode, in that it begins with an empty distribution tree and adds branches only as the result of explicit requests to join the distribution.

Rendezvous Point
– Instead of flooding the network to determine the status of multicast members, PIM-SM defines an RP. When a sender wants to send data, it first does so to the RP. When a receiver wants to receive data, it registers with the RP.
– When the data stream begins to flow from sender to RP to receiver, the routers in the path automatically optimise the path to remove unnecessary hops.

Automating Distribution of RP
Done via:
– Auto-RP
-Bootstrap Router (BSR)
– Anycast-RP

Configuring Internet Group Management Protocol (IGMP)
– Hosts use IGMP to dynamically register themselves in a multicast group on a particular LAN.
– Hosts identify group membership by sending IGMP messages to their local designated multicast router.

IGMP Snooping
Basically it’s a Layer 2 feature that makes sure only ports that have requested a multicast group receive them, and once that port has finished, or no longer sends keepalives, the port is pruned from the group.

In more detail, IGMP snooping requires the LAN switch to examine, or “snoop,” the IGMP join and leave messages sent between hosts and the first-hop designated multicast router.

Preparing the Campus Infrastructure to Support Wireless

Only interesting thing I read is that the only real configuration you need to do on Cisco switches, is configure the access point port as a trunk.

Preparing the Campus Infrastructure to Support Voice

Voice VLANs
Cisco switches offer a unique feature called Voice VLANs, alternatively named auxiliary VLAN. The voice VLAN feature enables you to overlay a voice topology on to a data network seamlessly. Voice VLANs provide for logical networks, even though the data and voice infrastructure are physically the same.

-If an end-user workstation is attached to the Cisco IP Phone that connects to a Cisco switch with a Voice VLAN configuration, traffic rom the user workstation is switched through the phone on the native VLAN, by default.
– The native VLAN is not tagged and is the actual switch port VLAN config. The Cisco IP Phone sends traffic with an 802.1q tag if a Voice VLAN is configured for a VLAN besides the native VLAN.

Basic IP Phone config:
S1(config)#mls qos
S1(config)#int fa0/0
S1(config-if)#switchport mode dynamic desirable
S1(config-if)#switchport voice vlan 700
S1(config-if)#mls qos trust cos
S1(config-if)#power inline auto
S1(config-if)#spanning-tree portfast

Phones download config files from TFP. Use option 150 on DHCP Server to specify TFTP Server location.