Quality Of Service QoS Computer Networks

In one word, Quality of Service (QoS) can be referred as efficiency. We define Quality of Service as “How well or efficiently data transmissions are taking place”.

So the question arises “How can we improve our network’s efficiency?” or “How can we improve our Quality of Service?”. It can be achieved by managing data traffic which results in reduced packet loss, latency, and jitter on the respective network. This is all done by setting up priorities for specific types of data that traverse through the network.

The enterprise networks or commercial networks (networks which connects all the users and the system connected through a local area network, to the applications in the data center) they provide services in the form of applications, for example, WhatsApp, from which we can send and receive various kinds of data, like audio, video, text, etc. Organizations like this use Quality of Service to meet the traffic requirements which results in the prevention of the degradation of the quality, which can be caused by packet loss, delay, and jitter.

How to achieve Quality of Service?
Let’s get into some details and say, your organization wants to achieve Quality of Service, which can be done by using some tools and techniques, like jitter buffer and traffic shaping.

Jitter buffer
This is a temporary storage buffer which is used to store the incoming data packets, it is used in packet-based networks to ensure that the continuity of the data streams doesn’t get disturbed, it does that by smoothing out the packet arrival times during periods of network congestion.

Traffic shaping
This technique which is also known as packet shaping, is a congestion control or management technique that helps to regulate network data transfer by delaying the flow of least important or least necessary data packets.

QoS is included in the service-level agreement when an organization signs it with its network service provider which guarantees the selected performance level.

There are 2 types of Quality of Service Solutions:
1. Stateless solution: Here, the server is not required to keep or store the server information or session details to itself. The routers maintain no fine-grained state about traffic, one positive factor of this is, that it’s scalable and robust. But also, it has weak services as there is no guarantee about the kind of performance delay in a particular application which we encounter. In the stateless solution, the server and client are loosely coupled and can act.

2. Stateful solution: Here, the server is required to maintain the current state and session information, the routers maintain per-flow state as the flow is very important in providing the Quality-of-Service which is providing powerful services such as guaranteed services and high resource utilization, provides protection, and is much less scalable and robust. Here, the server and client are tightly bounded.

Also Read: The OSI Model
Quality of Service Parameters:
QoS can be measured quantitatively by using several parameters

* Packet loss: it happens when the network links become congested and the routers and switches start dropping the packets. When these packets are dropped during real-time communication, such as audio or video, these sessions can experience jitter and gaps in speech.

* Jitter: occurs as the result of network congestion, timing drift, and route changes. And also, too much jitter can degrade the quality of audio communication.

* Latency: is the time delay, which is taken by a packet to travel from its source to its destination. For a great system, latency should be as low as possible, ideally, it should be close to zero.

* Bandwidth: is the capacity of a network channel to transmit maximum possible data through the channel in a certain amount of time. QoS optimizes a network by managing its bandwidth and setting the priorities for those applications which require more resources as compared to other applications.

* Mean opinion score: it is a metric for rating the audio quality which uses a five-point scale, with a five indicating the highest or best quality.

Implementing Quality of Service:
We can implement Quality of service through three of the following existing models:

1. Best Effort: if we are applying this model then, it means that we are prioritizing all the data packets equally. But since we all setting the priority order like this, then there is no guarantee that all the data packets will be delivered, but it will put up the best effort to deliver all of them. Point to remember is, that the best-effort model is applied when networks haven’t configured with the QoS policies or incase their network infrastructure does not support QoS.

2. Integrated Services: or IntServ, this QoS model reserves the bandwidth along a specific path on the network. The applications ask the network’s resource reservation for themselves and parallelly the network devices monitor the flow of packets to make sure network resources can accept packets. Point to remember: while implementing Integrated Services Model, the IntServ-capable routers and resource reservation protocol is necessary. This model has limited scalability and high consumption of the network resources.

3. Differentiated Services: in this QoS model, the network elements such as routers and switches are configured to serve multiple categories of traffic with different priority orders. A company can categorize the network traffic based on its requirements. Eg. Assigning higher priority to audio traffic etc.

We now know what Quality of Service is and how it can be implemented.

You may also like: