Chapter 7

Identifying Voice Networking Considerations

7.1. Explain How Applications and Traffic Flow Can Affect Network Design
Customers rate the convenience of an application by the length of time it takes to complete the
transaction. They also expect the application to be available whenever they want to use it.
Applications fast response time is considered Critical for the user. Applications considered critical
by organizations include Emergency services, Voice and video, monitoring and transmission etc.
The measurement of application performance should combine user satisfaction with normal
technical metrics, such as throughput, or the number of successful transactions.

7.2. Application Characterization

7.2.1. Application Characterization

The application characterization process involves looking at the following aspects of network
applications:

 How the applications work on the network

 The technical requirements of the application
 How applications interact with each other on the network

Designer determines which applications are considered business-critical. Traffic from different
types of applications results in varying network demands. The network designer recognizes four
main types of application communication.

 Client-to-client
 Client-to-distributed server
 Client-to-server farm
 Client-to-enterprise edge

On an existing network, the first step in characterizing applications is to gather as much

information about the network as possible. This includes gathering information from:

1|Page
• Organizational input: existing documentation about network and verbal input from customer
personnel

• Network audit: gathers information about network devices, monitors traffic, and reveals details
of the current network configuration

• Traffic analysis: provides information about how the applications and protocols use the network.
There are two kinds of traffics which are

 Internal traffic: identify areas where high bandwidth is needed, and possible bottlenecks
 External traffic: determine placement of firewalls and DMZ networks

Network-Based Application Recognition (NBAR) is Cisco utility that conducts audits and traffic
analysis. It is a classification engine that recognizes wide variety of applications.

Another tool is Cisco IOS NetFlow. NetFlow efficiently provides a set of services for IP
applications. Services include:

 Network traffic accounting

 Usage-based network billing
 Network planning
 Security
 Denial of Service monitoring capabilities

NetFlow provides valuable information about network users and applications, peak usage times,
and traffic routing.

7.3. Network devices

Types of hardware installed on a network affect performance of application. For example,

applications used for voice and video can be affected by hardware delays, causing performance
to degrade. Hardware delays can be caused by:

 Processing time that a router takes to forward traffic

 Older switches that are not able to handle traffic loads generated by modern applications
one way to ensure high performance is to use the top-down approach.

2|Page
  Top-down approach adapts the design of physical infrastructure to the needs of the network
applications
 Network devices are chosen only after a thorough technical requirements analysis

Transaction-Processing Applications

Transaction-processing is type of processing in which computer responds immediately to user

requests. Each request generated by user is a transaction. For example, consider what happens
when a customer purchases tickets online for an event at the sports stadium. This single transaction
generates all the following operations on the network:

 Web traffic from the client to the network

 Database transactions
 Customer order transaction
 Order processing transaction

Transaction-processing applications:

 Additional operations required

 Immediate response to user requests
 Redundancy and security required

Redundancy

Incorporating transaction applications requires the designer to consider the impact of each
transaction on the network. This process is crucial, because additional cabling or devices may be
needed to provide the redundancy or available throughput that these transactions require. Two
common protocols for implementing redundancy are:

 Rapid STP (RSTP): prevents Layer 2 switching loops that can occur with redundant
switches
 Hot Standby Routing Protocol (HSRP): can provide Layer 3 redundancy in the network.
HSRP provides immediate or link-specific failover and a recovery mechanism

3|Page
Redundant links and devices can be implemented in proposed network design at both Distribution
and Core Layers.

Real-time Applications

When designing the network to accommodate real-time applications, the designer must consider
how the network infrastructure will affect application performance. These considerations include
physical elements of infrastructure:

 Hardware devices and connections

 Network topology

7.4. Implementing QoS in Traffic Queues

Customers expect immediate access to their emails and to the files that they are sharing or
updating. To help ensure this availability, the network designer takes the following steps:

 Securing file and mail servers in a centralized location, such as a server farm.
 Protecting the location from unauthorized access by physical and logical security measures.
 Creating redundancy in the server farm that ensures that if one device fails, all files are not
lost.
 Configuring redundant paths to the servers.

7.4.1. Quality of service (QoS)

QoS refers to capability of network to provide preferential service to selected network traffic.
Primary goal of QoS is to provide priority, including dedicated bandwidth, controlled jitter and
latency, and reduced packet loss. Network designers must consider how QoS issues affect not only
the devices on a network but also the applications that use the network. Users perceive service
quality based on two criteria:

 The speed with which the network reacts to their requests

4|Page
  The availability of the applications they want to use QoS helps to manage these issues for
traffic flows within the network infrastructure and for the applications that use the network.
Some Cisco devices, such as routers, have built-in QoS mechanisms.

Figure 7.1 Priority Example

7.4.2. Implementing QoS in Traffic Queues

To implement QoS on a network, the designer follows three basic steps to ensure that traffic is
properly prioritized:

Step 1: Identify Traffic Requirements: Determine the QoS requirements needed for the different
types of traffic such as voice, mission-critical applications, and which low priority traffic can be
marked as best-effort.

Step 2: Define Traffic Classes: After traffic has been identified, it can be placed in appropriate
classes, such as voice traffic, which has highest priority, followed by mission-critical applications.
All other traffic can be prioritized as normal or low depending on the purpose of data. Packets are
marked to indicate the class to which they belong.

5|Page
Step 3: Define QoS Policies: The last step is to define the QoS policies to be applied to each class.
These policies include scheduling traffic queues and rules for managing congestion.
Where implement QoS: QoS can b

Where implement QoS: QoS can be implemented at the Access, Distribution, and Core Layers
of a network.

Layer 2 Devices

 Layer 2 switches at the Access Layer can support QoS based on IEEE 802.1p Class of
Service (CoS).
 The Layer 2 switch QoS uses classification and scheduling to prioritize sending frames
from the switch into the network.

Layer 3 Devices can support QoS based on:

 Physical interface, IP addresses, logical port numbers, and QoS bits in the IP packet
 QoS in Distribution and Core Layer devices must be supported in both directions of
traffic flow.

6|Page
Some key considerations for designing a voice network:

1. Bandwidth: Voice traffic requires a certain amount of bandwidth to ensure high-quality calls.
The amount of bandwidth required will depend on the number of concurrent calls and the codec
used to compress the voice data.

2. Quality of Service (QoS): QoS is critical for ensuring that voice traffic is prioritized over other
types of traffic on the network. This helps to prevent delays, jitter, and dropped calls.

3. Redundancy: Voice networks should be designed with redundancy in mind to ensure high
availability. This may include redundant servers, switches, and routers, as well as backup power
supplies.

4. Security: Voice networks are vulnerable to various types of attacks, including eavesdropping,
denial of service (DoS), and toll fraud. Security measures should be implemented to protect against
these threats.

5. Scalability: Voice networks should be designed to accommodate future growth and expansion.
This may include adding additional servers, switches, and routers, as well as increasing bandwidth
capacity.

6. Interoperability: Voice networks may need to interoperate with other systems, such as video
conferencing or unified communications platforms. Compatibility and integration should be
considered during the design phase.

7. Call routing: Call routing is an important consideration for voice networks, particularly for
organizations with multiple locations or remote workers. The network should be designed to
support efficient call routing and management.

8. Disaster recovery: Voice networks should have a disaster recovery plan in place to ensure that
calls can be rerouted in the event of a network outage or other disruption.

9. User experience: The user experience is critical for voice networks. The network should be
designed to provide high-quality, reliable voice calls with minimal latency and other issues that
can impact call quality.

7|Page