Quality Of Service or QoS (colloquial/everyday use). used to regulate traffic over ATM network devices, over Ethernet networks, using QoS Protocols and/or the IEEE 802.1P standard. ### Index ### ---- ! Introduction Standard [Internet] Protocol (IP)-based networks provide "best effort" data delivery by default. Best-effort IP allows the complexity to stay in the end-hosts, so the network can remain relatively simple. This scales well, as evidenced by the ability of the [Internet] to support its phenomenal growth. As more hosts are connected, network service demands eventually exceed capacity, but service is not denied. Instead it degrades gracefully. Although the resulting variability in delivery delays (jitter) and packet loss do not adversely affect typical Internet applications--email, file transfer and Web applications- other applications cannot adapt to inconsistent service levels. Delivery delays cause problems for applications with real-time requirements, such as those that deliver multimedia, the most demanding of which are two-way applications like telephony. Let's say a client on the network is experiencing poor audio quality with IP telephony. If there isn't enough raw capacity, you could just throw bandwidth at the problem. But many problems are more complex. For example, acceptable IP telephony requires a maximum 300-ms end-to-end delay. If you have seven 50-ms router hops from origin to destination, all the bandwidth in the world won't make the packets travel faster. In order to deliver QoS, you'll need to remove some hops, make the routers handle all packets faster or give those telephony packets some form of priority that gets them through the system faster. So in an effort to 'deliver' a certain standard of availability, there are specific protocols that can be used to restrict traffic flow or give specific network protocols priority over other traffic. ---- ! Implementation The primary selection criteria for a good network is available bandwidth. Bandwidth, of course, is measured in how many bits per second (usually measured in Mbits) are available for a connection. Quality of Service is implemented in order to sustain the performance of the network. Performance is typically measured in metrics like maximum end-to-end response time (how long it takes a packet to go from node A through all the switches, routers, firewalls and cables to node B); allowable packet-loss rate (how many packets can be dropped before the connection is deemed unacceptable), and jitter (variance in packet delivery delays). Once those QoS criteria are defined, then you can go to work to meet those needs. QoS does not create bandwidth, but manages it so it is used more effectively to meet the wide range or application requirements. The goal of QoS is to provide some level of predictability and control beyond the current IP "best-effort" service. The prospect of such a potentially drastic change makes many of the Internet's architects very nervous. To avoid these potential problems as QoS protocols are applied to the Net, the end-to-end principle is still the primary focus of QoS architects. As a result, the fundamental principle of "Leave complexity at the `edges' and keep the network `core' simple" is a central theme among QoS architecture designs. This is not as much a focus for individual QoS protocols, but in how they are used together to enable end-to-end QoS. ---- ! Protocols Applications, network topology and policy dictate which type of QoS is most appropriate for individual 'flows' or aggregates. To accommodate the need for these different types of QoS, there are a number of different QoS protocols and algorithms. such as: * [ResourceReservationProtocol Resource Reservation Protocol] ([RSVP]) : Provides the signaling to enable network resource reservation (otherwise known as Integrated Services). Although typically used on a per-flow basis, RSVP is also used to reserve resources for aggregates. * [DifferentiatedServices Differentiated Services] (DiffServ) : Provides a coarse and simple way to categorize and prioritize network traffic (flow) aggregates. * [MultiProtocolLabelingSwitching Multi Protocol Labeling Switching] (MPLS) : Provides bandwidth management for aggregates via network routing control according to labels in (encapsulating) packet headers. * [SubnetBandwidthManagement Subnet Bandwidth Management] (SBM) : Enables categorization and prioritization at Layer 2 (the data-link layer in the OSI model) on shared and switched IEEE 802 networks. * [802.11e] * [802.15.3] ---- ! Hardware [AsynchronousTransferMode ATM], commonly used in wide-area networks and enterprise backbones, but too expensive and complicated for most LANs, has outstanding built-in QoS features, but some of these features are not used extensively, only as required. To implement QoS over Ethernet, you need an additional protocol called [ResourceReservationProtocol Resource Reservation Protocol] (RSVP). Although RSVP doesn't offer all the QoS features of ATM, it does help reserve bandwidth to support real-time jitter-intolerant applications. The challenge is that RSVP is still new, and many devices and applications don't know how to use it yet. ---- ! Wireless Wireless networks, with a good response time, unreliable bandwidth, and unreliable latency and unreliable packet-loss prevention have a lower standard of QoS goals in comparison to ATM or ethernet. Goals to provide an acceptable jitter and routing response over PointToMultiPoint access.