| Lesson 5 || Enhancing performance with QoS |
| Objective || Define how QoS enhances network performance. |
Enhancing performance with QoS
Data traffic routed across an IP network might have to travel through a series of routers. By default, each router in the path handles the traffic on a first-come, first-served (best-effort) basis. Time-sensitive streams of data such as video streaming, real-time audio, and video conferencing require a regular flow rate across the network, and will not work properly if their packets have to wait in line at a busy router.
The Windows 2000 implementation of QoS provides selectable bandwidth reservation and priority levels for traffic flows. QoS provides a unidirectional mechanism by which you can manage network resources, whether on the local LAN segment, the core network, or the WAN--to provide the required service levels. QoS can run on any network topology that supports TCP/IP.
The following Slide Show explains when designers can use QoS.
- You can use QoS to reserve bandwidth for particular users, services, or applications
- You can use QoS to prioritize access to bandwidth, based on user needs.
- You can use QoS to partition bandwidth between best-effort traffic and higher or lower priority traffic
- You can use QoS to prevent non-adaptive protocols (such as UDP) from overusing network resources
All routers in the path between a sender
and receiver must be able to support QoS, and must validate the bandwidth requirement before a QoS session can be established. Include QoS in your design if:
- Bandwidth allocation is required
- QoS is supported by your planned applications
- All intermediate routers support Resource Reservation Protocol (RSVP)
As a network architect, you need to consider QoS in your design to support high-bandwidth applications in the face of limited available bandwidth on your network. Some applications are highly tolerant of network delays, and function normally within such a context. For example, SMTP mail is quite fault tolerant in the face of high network utilization, and SMTP messages will be delivered without error even when the network bandwidth is over-utilized.
However, if the organization that you are working with plans to implement extensive teleconferencing for mission-critical meetings, you will run up against the limitations of an Ethernet 10/100 network relatively quickly. In this situation, you should include an implementation of QoS in your design document so that the required amount of bandwidth for audio-conferencing applications is available when required. With QoS, you can assign the bandwidth to individuals or group members to ensure that bandwidth is not wasted on non-priority users or groups.
The goal of Internet of Things (IoT) is to bring electronic components online, thereby creating a volume of data that can be used with the existing computing and networking technologies.
The cloud isn't ideal for rapidly expanding IoT environmental requirements.
(FC) Fog computing moves some portion of the computing load (related to real-time services) from the cloud into edge fog devices.
FC is expected to become the subsequent major computing transition and this one has ability to overcome existing cloud limitations.
However the key obstacles facing Fog computing are:
- wide distribution,
- isolated coupling,
- quality-of- service (QoS) regulation,
- adaptability to conditions, and
- the standardization and normalization is still in phase of development.
Programming characterized organizing (SDN) will assist with tackling these hindrances.
SDN implies bound together organization control plane (which is isolated from information plane), permitting the presentation for cutting edge traffic signal and the arrangement instruments of organizations and assets.
On the grounds of SDN idea, later consolidating it with FC, the organization type can be adjusted to determine that multitude of cloud disadvantages, further developing IoT framework's within the context of QoS.
Engineering is created through the blend of autonomously investigated areas of SDN and FC to improve the QoS in an IoT framework. A calculation (which is subject to parcel the SDN essentially) is introduced to help the engineer, whose design is to choose the ideal passage and ideal spot to handle the information. The primary goal of this calculation is to give further developed QoS by apportioning the corresponding gadgets through the SDN regulator. A utilization case subject to the introduced design and calculation is then given and evaluated.
This utilization case or QoS boundary values (network use, cost, dormancy and power utilization) utilizes the iFogSim test system. The outcome demonstrates a significant improvement of the referenced QoS boundary values in the organization of haze with SDN. Also, once contrasted with an overall previous indistinguishable use case, the discoveries show further developed outcomes for power utilization, network use and dormancy.
When contrasted with a previous indistinguishable use case, the result of this paper shows multiple times less inactivity and twice less organization utilization. At long last the ground (IoMT, Industry 4.0, Green IoT, and 5G) that is impacted by this QoS improvement is extensively outlined in this paper.
The next lesson looks at QoS connections.