4.5 Network Capacity Planning

Recommend optimization for network capacity planning, such as topology and configuration changes, and QoS based on data interpretation

Network Assurance engineers must be proactive in monitoring network capacity and performance to ensure a consistently high-quality user experience. As organizations evolve, the ability to gather and interpret relevant data becomes increasingly important for making informed optimization decisions. This section will strengthen your understanding of key concepts and tools related to capacity planning and optimization, preparing you to effectively analyze data and recommend appropriate actions.

Key Concepts

Capacity Planning

Capacity planning is a crucial aspect of network assurance. It involves forecasting bandwidth requirements based on user needs and growth trends. This proactive approach ensures that users consistently receive the best possible digital experience over the network.

ThousandEyes WAN Insights can assist with capacity planning by integrating with Cisco Catalyst SD-WAN Manager. This integration obtains the maximum bandwidth for each circuit, providing valuable data for optimization decisions. For more information, refer to the product documentation on How Capacity Planning works with WAN Insights.

Capacity planning is essential for all traffic types but is particularly critical for applications sensitive to bandwidth limitations and network issues, such as video conferencing and Voice over IP (VoIP) calls.

Quality of Service (QoS)

Quality of Service is a key feature that can be leveraged alongside a robust capacity planning strategy. QoS helps prioritize network traffic to ensure optimal performance for critical applications.

To understand QoS, consider this analogy:

Imagine your network as a highway full of vehicles, where each vehicle represents a chunk of data. QoS acts like a traffic management system, ensuring that high-priority vehicles (critical data) don't get stuck in congestion.

Importance of QoS

Different network activities have varying requirements for optimal performance:

VoIP calls require low latency to avoid awkward pauses.
Video streaming needs a steady data stream to prevent buffering.

QoS ensures these activities run smoothly, even during peak network usage.

How QoS Works

Data packets are tagged with priority levels.
Network devices read these tags and sort data into different queues.
Higher priority data is processed first.
Some QoS implementations reserve bandwidth for critical applications, similar to a dedicated lane on a highway.

Data Interpretation for Optimization

To make informed optimization recommendations, you'll need to analyze data from various sources, including:

Telemetry
SNMP (Simple Network Management Protocol)
CLI (Command Line Interface) outputs
Syslog messages
NetFlow

These data sources provide insights into network performance, utilization, and potential bottlenecks. By interpreting this data, you can identify areas for improvement and suggest appropriate optimizations.

Resources

Sample Questions

4.5 Question 1

You're analyzing NetFlow data for a network supporting voice and video traffic. The data shows consistent spikes in delay and jitter during peak hours. Which optimization would you recommend?

A) Implement a complete QoS redesign
B) Increase bandwidth on all network links
C) Tune the existing QoS configuration to prioritize voice and video traffic
D) Replace all network hardware with newer models

4.5 Question 2

SNMP data indicates that a wireless access point is experiencing high channel utilization and increased retransmissions. What optimization would you recommend to improve voice call quality for users on this access point?

A) Increase the transmit power of the access point
B) Change the access point to a different, less congested channel
C) Disable all non-voice traffic on the wireless network
D) Implement strict admission control for all wireless clients

4.5 Question 3

CLI outputs show that a router's egress queue for voice traffic is consistently full, leading to increased latency. Based on this data, which optimization would you recommend?

A) Increase the queue size for voice traffic
B) Implement traffic shaping on non-voice traffic
C) Disable QoS on the router to allow all traffic equal priority
D) Replace the router with a higher-capacity model

4.5 Question 4

The following exhibit shows the Capacity Planning results for a router interface connected to an ISP, which provides a 1Gbps connection: Based on the evidence, which action is most likely to fix the observed behavior?

A) Request a link increase from the ISP
B) Reconfigure maximum capacity for the interface
C) Restrict the Web Sites that can be visited from the site
D) Reconfigure business hours settings