1. Chapter 4
Network Performance and Optimization
Identify and Resolve Network Problems EIS HNS3 05
1221
TVET - IT Support Services- 2016 Entry Regular
2. Network Performance & Optimization
Network performance refers to the quality and
speed of data transmission across a network.
Network Optimization ensures efficient, reliable
communication for users and applications.
Key metrics: latency, jitter, bandwidth,
Throughput, Error Rate, and packet loss.
3. Detecting Network Issues
Common Problems:
Congestion: Excessive data on the network causes delays
and packet drops.
High Latency: Slow response times, especially noticeable
in VoIP or online gaming.
Packet Loss: Data packets fail to reach the destination,
disrupting communication.
Slow Transfers: Data (files, videos, backups, etc.) moves
across a network or between devices at significantly lower
speeds than expected.
Detecting Tools:
Ping, Traceroute, NetFlow, SNMP
4. Network Metrics
Latency : Like waiting for someone to respond in a phone call. When you
speak, there's a small delay before the other person hears you. In networks,
it's the time it takes for data to travel from your device to its destination and
back.
Throughput: Similar to the speedometer in your car. Instead of measuring
miles per hour, it measures how much data can move through the network
in a given time (like megabytes per second).
Packet Loss: Imagine sending letters in the mail. Sometimes letters get lost
along the way. Packet loss happens when small chunks of data (called
packets) don't reach their destination. Just like you'd need to resend a lost
letter, the network needs to resend lost packets.
Jitter: Picture watching TV - sometimes the broadcast might stutter or
freeze briefly. Jitter is like that stuttering effect in data transmission, where
packets arrive irregularly rather than in a smooth flow.
5. Cont.
Bandwidth: Like the number of lanes on a highway.
Just as more lanes let more cars pass through at once,
bandwidth determines how much data can flow
through the network simultaneously.
Error Rate: Like counting mistakes in a document. It
tracks how many problems occur during data
transmission compared to the total amount of data sent.
Response Time: Like ordering food at a restaurant - it's
the total time from placing your order until getting your
food. In networks, it measures how long it takes for
your request to be processed and get a response back.
6. Core Network Optimization Metrics
METRIC DEFINITION IMPACT ON OPTIMIZATION
Latency
Time for data packets to travel
from source to destination
Critical for real-time applications and
user experience
Throughput
Volume of data transferred per
unit time
Determines network capacity and
efficiency
Packet Loss Failed packet deliveries Affects service quality and reliability
Jitter Variation in packet arrival times
Impacts real-time communication
quality
Bandwidth Maximum data transfer capacity
Determines concurrent operations
support
Error Rate
Ratio of errored bits/packets to
total sent
Indicates hardware/connection issues
8. Quality of Service (QoS)
Prioritize certain types of traffic (e.g., VoIP, video
conferencing).
QoS Techniques:
Traffic classification and marking (DSCP, CoS)
Bandwidth reservation
Queuing mechanisms: first in first out(FIFO),
weighted fair queuing(WFQ)
9. Traffic Shaping and Bandwidth Management
Regulate traffic flow to prevent congestion.
Methods:
Rate limiting: Restrict bandwidth usage for
specific applications.
Scheduling: Delay lower-priority traffic to
prioritize business-critical apps.
Tools: Router configurations, software-based tools
(e.g., pfSense, Cisco IOS).
10. Load Balancing
Distribute traffic across multiple servers or links
to optimize resource use.
Types of load balancing:
Round-robin: Equal distribution.
Least connections: New traffic sent to the least-
burdened server.
Geographic: Based on user location.
Benefits: Increased availability, fault tolerance,
and scalability.
11. Round-robin
Client Load Balancer
→
├─ Request 1 Server A
→
├─ Request 2 Server B
→
├─ Request 3 Server C
→
└─ Request 4 Server A (cycle repeats)
→
Least connections
Client Load Balancer
→
├─ Server A (5 active connections) Skipped
→
├─ Server B (2 active connections) Chosen (least busy)
→
└─ Server C (10 active connections) Skipped
→
Geographic
User in Europe Load Balancer Nearest Server: Germany
→ →
User in Asia Load Balancer Nearest Server: Singapore
→ →
User in USA Load Balancer Nearest Server: Virginia
→ →
12. Preventive Maintenance
Practices:
Regular firmware/software updates for network
devices.
Cable testing and equipment cleaning.
Scheduled performance audits and log reviews.
Benefits: Reduce unexpected downtimes and
extend equipment lifespan.
13. Network Performance Tuning
Adjust MTU (Maximum Transmission Unit) Settings
Ensure packets are optimally sized to minimize fragmentation and
improve throughput.
Optimize Routing Tables & DNS Configurations
Streamline routing paths to reduce latency.
Ensure DNS resolves quickly to prevent delays in network
communication.
Monitor Interface Utilization & Adjust Thresholds
Track bandwidth usage and adjust load balancing or QoS settings as
needed.
Administrative Task:
Continuously Refine Configurations
Regularly review performance metrics and adjust settings based on
real-world feedback.
