1.1.1 Introduction to “Quality of Service (QoS)” in Networks
A brief introduction of “Quality of Service (QoS)”. The term QoS refers to the ability of a network to provide a better service. It is a service that selects network traffic over various technologies and IP-routed networks that may use any of these underlying technologies. Some examples of those technologies are stated below:
· Frame Relay
· Asynchronous Transfer Mode (ATM)
· 802.1 networks
· Synchronous Optical Network (SONET)
The main goal of “Quality of Service (QoS)” is to provide priority on dedicated bandwidth, controlled jitter and latency, and improved loss characteristics. Also, it is important to make sure that providing priority for a single or more flows will not make other flows fail. QoS technologies provide the fundament blocks that are used for future business applications in campus, Wide Area Network (WAN), and service provider networks.
In today’s enterprise IT infrastructure having a proper “Quality of Service (QoS)” into a IP network is an extremely main factor. QoS is a necessity for voice and video streaming in the network, it is also a major factor for it supports the growing “Internet of Things (IoT)”. In this section of this article an explanation of the importance of “Quality of Service (QoS)”, concept behind it and state some case scenarios on how it benefits our user experience.
1.1.2 Importance of “Quality of Service (QoS)”
In applications that are running in the network which are sensitive to lag. These applications similarly use the “User Datagram Protocol (UDP)” as disagrees with the “Transmission Control Protocol (TCP)”. The main comparison between TCP and UDP is that it has a connection to time sensitivity. TCP will transmit old packets which are lost in transit, as for UDP does not. For which a file to transfer from a single PC to another PC, TCP will be used, for if any packets are lost or arrive out of order, TCP protocol has the capability to transmit old packets and redeliver the packets to create the old file on the destination PC.
Anyhow for “User Datagram Protocol (UDP)” provisions for example, such that an IP telephone call, at whatever bundle reduction can’t be retransmitted because of those voice packets that come in. Likewise, a requested stream to re-transmitting those lost packets will be futile. Due to that whatever delayed packets to applications running on UDP wind up having problems. Take voice calls for an example, losing even a couple packets may end up having the voice quality becoming static in a way transferring of communication won’t sound good. In a way packets are quite sensitive to jitter. Jitter is defined as a type of delay in a media application.
If a network has a lot of bandwidth and has no traffic that may bursts above on what it can handle, there won’t be a problem with packet loss, delay or jitter. But for many or a lot of enterprise networks, there will be times in where connections become overly congested to the point where routers and switches start dropping packets because they are coming in and out faster than what the hardware is capable of processing. If that is the case streaming applications may end up suffering due to overloading or traffic and this is where “Quality of Service (QoS)” comes in.
1.1.3 How “Quality of Service (QoS)” Works
In a network infrastructure “Quality of Service (QoS)” manages packet loss, delay and jitter. When handling on a vast number of bandwidth, the first thing to do would be to identify in what way the application would benefit from handling packet loss, delay and jitter. Once a network and administrators of the application identify the application that would need priority over a bandwidth within a network the next step would be to identify the said overloading or traffic in the network. There are many ways to identify traffic in a network. Some examples would be “Class of Service (CoS)” and “Differentiated Services Code Point (DSCP)”. In CoS a data stream will be marked in a frame header of layer 2 while DSCP a data stream will be marked in a frame header of layer 3. There are various applications that may identify traffic in a network and can mark it differently, which does allow the network hardware to categorize data into different types or categories.
Now that the data streams are arrange under separate groups, use that information to place a configuration or setting in those types in order to provide proper treatment of some data streams than others. This concept is well known as queuing. An example would be if a tagged voice traffic and a policy is made to give access to the all of the network bandwidth within a link, the routing or switching of a certain hardware will end up moving those packets to upfront or ahead of the queue and transmit them instantly. But if a common configuration of TCP data transfer stream is marked tagged with a low priority, it will be on standy until there is enough bandwidth for it transmit with. If the queues pile up in insurable amounts, the low priority packets end up being the first to be dropped by the system.
1.1.4 “Quality of Service (QoS)” Sample Case Scenario
As mentioned on the previous sections, the most common used cases for “Quality of Service (QoS)” are media streams in a network. But more examples, especially now that “Internet of Things (IoT)” is starting to get popular or start off. An example is in the industrial sector, wherein machines are starting to hold the network to provide actual time status of data on any type of problems that may end up occurring. Any cost of delay in the finding the issue in the network can cause in manufacturing mistakes which may cost a lot of cash in each second. By applying QoS in the network, the manufacture process status data stream takes priority within the network that ensures information flows in an acceptable manner. Another case may be in the streaming of different types of smart sensors for huge IoT plans such as a smart building. Most of the information gathered and studied, such as temperature, humidity, and location awareness, are highly sensitive in time. Due to this sensitivity, the type of data must be properly found, marked and queued appropriately. It is safe to say that a network connectivity must continue to undergo change into every aspect of all sorts like business stuff, for “Quality of Service (QoS)” does play an increasingly vital role in insuring that a particular data streams are given priority than others in order for the network to function efficiently.
1.2.1 “Quality of Service (QoS)” in Wired Networks
Before anything else it is best to define a wired network. A wired network is a common type of network which has wired configurations. It is a network that has a lot of wirings that connects from one node to another node. Most wired networks use Ethernet cables to transfer data between connected computers within a network. In a small wired network, one router or switch may be used to connect all the computers. Larger networks often involve multiple routers or switches that connect to one another. Some of these devices typically connects to a cable modem, T1 line, or other type of Internet connection which provides Internet access to all of the devices that are connected within the network. In terms of “Quality of Service (QoS)” it enables a network administrator to guarantee a minimum bandwidth for certain types of traffic & limit the maximum bandwidth for other types of traffic that overload the network. As mentioned on the previous section QoS enables real time applications such as video and voice to maintain low latency, jitter levels which are absolutely critical to attain the best user experience and they minimize the effects of any packet loss and buffer overflow in networks that end up being congested. Additionally, QoS can also play a role in mitigating attacks in a system like “Denial-of-Service (DoS)” or “Distributed Denial-of-Service (DDoS)”.
1.2.2 Concept behind “Quality of Service (QoS)” in Wired Networks
Based from the article “Quality of Service (QoS)” parameters may be configured in network switches and routers that are manageable (Rajesh, 2011). Following what was stated there are possibilities that all the packets or frames within the network cause traffic that are entering into each switch or router which are classified into various “Class of Service (CoS)” categories. So that the network protocol and management traffic may as well be in one CoS category, voice and video traffic may be another category in CoS, and many more. But before that, all the network traffic should be classified appropriately and marked as close to the source end point as much as possible. Once is is classified, they have to be Queued using an appropriate hardware queues based on their which CoS category they are in at all levels within the network. Network traffic of individual packets may as well be classified and grouped using one of the following methods:
· Explicit 802.1p or DSCP marking
· VLAN / Switched port based grouping
· MAC address based grouping
It is important to ensure that a single scheme of “Quality of Service (QoS)” is implemented uniformly throughout the network. In the upcoming sections of this paper the shall be discussed closely as Hardware Queuing and Explicit 802.1p or “Differentiated Services Code Point (DSCP)” marking which are used more commonly. Hardware Queuing is a concept were within a switch or a router within a network, there is a possible potential problem of buffer overflow because of the congestion within the network. So that there is a need of separate hardware queues that may be implemented for sorts of different categories of network traffic that are “Class of Service (CoS)” types and packets which can be transmitted independently and in accordance to the level of priority of different types of traffic flow. A network switch in general supports ‘n’ number of hardware queues per each physical port, for an example four hardware queues per port in a switch hardware. Base on that each hardware queue can be programmed with appropriate bandwidth prioritization and limitation parameters in order to apply “Quality of Service (QoS)” in the network.
1.3.1 “Quality of Service (QoS)” in Wireless Networks
Wireless frameworks are ending up more inescapable starting late for it is transforming into an example, it ranges from versatile straightforward and modernized cell correspondence up to satellite telecom till now. Thusly, having or supporting a high “Quality of Service (QoS)” in passing on voice, video and data in remote frameworks has ascended as a champion among the most basic trial of today. Around there of this article, we should discuss indisputably the hugest subjects stressed over QoS in remote frameworks, focusing on perspectives like survivability, design and Ad hoc composes.
1.3.2 “Quality of Service (QoS)” Wireless Networks Survivability
In light of the article, ‘network survivability’ was utilized as a part of the past to signify the request of keeping up usefulness as a result of a system segment disappointment for example, the evacuation of a connection or the interruption of an office area (Bil`o, 2003). It is surely pleasing what was expressed, for it is fundamental to avert outrageous awful circumstances wherein as an outcome of that part disappointment, the “Quality of Service (QoS)” inside the system is influenced by it. Among the others, an approach which is emerging in the last period for addressing the ‘network survivability’ is based on a two-levels view of the network itself that are stated below which are mentioned by the article:
· The “working network” which is the backbone where communication is carried out in the absence of failures in the network, that are designed by optimizing any given function of objective.
· The “emergency network”, which is a set of links that are inactive and ready to be used to create alternative routes as soon as the working network undergoes possible component failure which is affecting its purpose.
As expressed over, a emergency network should be as close as could be expected under the circumstances, both as far as physical parts and of interchanges convention which are an essential system, that is ignoring it to some degree, in which the request of the maximal productivity as for the streamlining capacity that is tended to by the essential system. The inspiration driving this sort of approach is that one may like to pay regarding proficiency, once the exchanging into the crisis arrange is quick and effortless. This is much more evident at whatever point the disappointment in which influenced the system is potentially transient, and after that an opportunity to expect that it changes back to the essential one quickly and in a convenient way.
1.3.3 “Quality of Service (QoS)” Wireless Networks Layout design
The combination of wireless and ” Asynchronous Transfer Mode (ATM)” or ” Wireless (WATM)” systems is rising as a standout amongst the most encouraging methodologies ready to help clients’ portability while keeping up the nature of administration offered by the established ATM. This mix happens at various levels and yields diverse situations, for example, End-to-End WATM and WATM Interworking, connected separately to make new remote systems with ATM virtual channels stretching out until the versatile terminals and at a more outside level for interconnecting distinctive existing remote subnets.
In view of a specific article it expressed that the standard ” Asynchronous Transfer Mode (ATM)” designs must be developed with a specific end goal to make an alternate virtual circuit for each station (Fioravanti, 2003). It must be distinctive additionally other than the bounce check and the heap, their execution is likewise assessed by methods for a further assessment in parameter. The virtual channel removes which measures the time that is expected to reproduce virtual diverts amid hand-off stages. that is when versatile terminals switch between adjoining stations. All the more absolutely the separation that is between virtual channels of two neighboring hubs is equivalent to the quantity of virtual ways that must be erased and added to one virtual circuit with the end goal for it to acquire the other one. With a specific end goal to influence the rerouting to stage unnoticeable to clients and accordingly to get nature of administration that is palatable, the most extreme separation between the two virtual channels must be kept up as low as could be allowed. It must be a characteristic combinatorial issue that emerges in which appropriate exchange offs has be resolved between the diverse execution measures.
1.3.4 “Quality of Service (QoS)” Ad hoc Wireless Networks
Wireless ad hoc networks may as well be called as wireless multi-hop networks, are formed by multiple nodes. Nodes were each of them are possessing a wireless transceiver, communicating amongst themselves. An ad hoc network may be used to exchange information between the nodes and to allow nodes to communicate with remote or rural sites that they otherwise would not have the capability to reach. Wireless ad hoc networks may be either static such as sensor networks, or mobile such as “Unmanned Aerial Vehicle (UAV)” networks. The most important design criterion for any type of networks is a network that guarantees the “Quality of Service (QoS)”.
There is an article that mentions mobile ad hoc networks, in which nodes have to be balance for a variety of tasks including sensing and communications (Pandya and Pottie, n.d.). Based on what was mentioned, nodes might change location or trajectory due to sensing purposes, with limiters on disruption to the “Quality of Service (QoS)” of the network. For example, a sensor node might be needed to have its position changed for an improvement in source identification without disrupting the existing communications. In a way “Unmanned Aerial Vehicle (UAV)” networks, has a node that might have to move to replace the failed backbone node.
An article stated that Ad hoc networks are connected by wireless links that are mobile devices and are self-configuring networks. Devices which are able to move by themselves in an arbitrary fashion. It also mentioned that Ad hoc networks are used in situations where an infrastructure is unavailable to deploy (Minh Do, Landmark, and Kure, 2010). Which is true and for that an example would be emergency operations, relief operations, temporary networks, transient vehicle-to vehicle communication and tactical military networks. In most of these settings multipoint-to-multipoint multimedia communication is a main requirement for it to work. Multicast is said to be an efficient method to implement multipoint-to-multipoint communication. Multimedia content which is mentioned on the previous sections really requires networks with “Quality of Service (QoS)” for it to function smoothly.
Base on some of the articles previously mentioned “Quality of Service (QoS)” is a term that refers to the ability of a network to provide a better service. The primary goal of QoS is to provide priority on dedicated bandwidth, controlled jitter and latency, and improved loss characteristics. Also, it is important to make sure that providing priority for a single or more flows will not make other flows fail. It is best to configure QoS properly because if a network has a lot of bandwidth and has no traffic that may bursts above the limits on what was configured, there won’t be a problem with packet loss, delay or jitter. But for a lot of enterprise networks, there will be times in where connections become over capped to the point where routers and switches start losing packets for they are coming in and out faster than whatever hardware is capable of processing.
In the upcoming sections of this paper, advantages of wired networks and wireless networks will be tackled. Hardware or devices needed will be discussed in-depth. The architecture or design that will be needed to compare the “Quality of Service (QoS)” of both wired and wireless networks in respect to the main topic of this article. Also, a wired topology will be compared on a wireless ad hoc which is the main topic of this paper. Further more in-depth as to why QoS is vital in a network will be tackled more in the upcoming chapters.