The Need for Quality of Service
Converged networks combine various types of traffic, such as voice, video, and data, which traditionally use separate dedicated networks. However, with the network improvements over the years, adapting to new technologies, standards, and approaches, and mixing these types of traffic, this blending presents unique traffic characteristics and challenges.
The issues within converged networks typically arise due to the coexistence of constant, small-packet voice flows alongside bursty video and data flows. Voice and video traffic is time-sensitive, requiring real-time delivery and predictability, while most data traffic can tolerate delays. Therefore, critical traffic, like voice and real-time video, can suffer and create various issues without proper prioritization.
The diverse nature of these traffic types can lead to quality problems. Data traffic exhibits unpredictable patterns and arrival times. Its importance varies, from non-interactive, non-delay-sensitive applications like email to mission-critical functions. Voice traffic, on the other hand, relies on constant bandwidth and predictable packet arrival times, while video traffic comprises different subtypes with varying bandwidth requirements, delay tolerances, and loss tolerances.
Managing these differing traffic flows and prioritizing critical ones is essential to prevent network congestion and ensure a quality experience for end users.
Quality Issues in Converged Networks
While converged networks are efficient, they can lead to challenges like competition for bandwidth and prioritization of different traffic types, affecting the quality of communication and data exchange. Quality issues on converged networks arise from four main problems, collectively impacting the quality of service on converged networks:
- Bandwidth Constraints: The network's capacity can be strained by large files, multimedia content, and the growing use of voice and video. This competition for limited bandwidth often leads to insufficiencies, requiring more bandwidth than available.
- Delay: Delay refers to the time it takes for a transmitted packet to reach its destination. This encompasses variable factors like processing and queueing delays and fixed elements like serialization and propagation delays, all contributing to end-to-end delay.
- Jitter: Jitter signifies variations in latency or time for a signal to travel from sender to receiver. It can disrupt the smooth flow of data packets as they traverse the network.
- Packet Loss: Loss of packets typically results from congestion, faulty connections, or malfunctioning network equipment.
Mitigating Quality Issues in Converged Networks
Addressing network quality issues involves managing limited bandwidth, minimizing delay, compensating for jitter, and preventing packet loss. Various techniques can be employed to ensure smooth data transmission and a seamless user experience:
- Bandwidth Constraints: Limited bandwidth results from network congestion or lower-bandwidth segments. Boosting link capacity is ideal but not always feasible to manage congestion due to cost or time constraints. Alternatives include queuing to prioritize traffic or using compression to reduce transmitted bits.
- Delay: Delay can be tackled by increasing link bandwidth, prioritizing critical traffic through queuing, or reducing transmitted bits via compression.
- Jitter: Jitter compensation involves de-jitter buffers in endpoints like IP phones or video gateways. These buffers smooth out packet delivery, adding delays but ensuring smoother real-time traffic. Excessive jitter leads to dropped packets and reduced media stream quality.
- Packet Loss: Packet loss occurs when router output queues run out of space, known as tail drop. Solutions involve increasing link bandwidth, queuing to allocate buffer space, or preventing congestion through shaping or packet dropping.