International telecommunication union
QoS classes and their performance objectives
Download 0.91 Mb.
|
- Navigate this page:
- Allocation guidance
- Methods for QoS Budget Allocation
- Static Approach
- Pseudo-static Approach
- Signalled Approach
- Impairment accumulation Approach
- Need for a new approach
QoS classes and their performance objectivesQoE requirements of the applications/services for IMT-2020 can be identified by an extremely wide range from best effort to very stringent level. These various levels should be classified as follows;
Specific QoE classes and their performance objectives for IMT-2020 are for further study. Editor’s Note: This is an important topic and we are more than welcome to have suggestions
IMT-2020 will consist of various use cases that will require different connectivity configurations. This means that the QoS framework has to take into account various cases in order to ensure QoS of connectivity in IMT-2020. One of the aspects that deserve special treatment is QoS budget allocation (also known as impairment allocation). Although end-to-end QoE requirement will be defined, the implementation will be different for various networks with different circumstances. This calls for an in-depth study on budget allocation approaches. In this context, this section outlines different methodologies for QoS budget allocation and identifies additional consideration to be considered in the IMT-2020 QoS framework.
This section will outline QoS budget allocation methods as described in ITU-T Recommendation Y.1542 and identify the need for a new approach for IMT-2020 specific connectivity cases. If details are necessary for the first four subsections, please consult Y.1542.
This approach divides the UNI-to-UNI path into a fixed number of segments and budgets the impairments such that the total objective is met in principle. It requires that individual segments have knowledge of the distance and traffic characteristics between the edges of their domains, as these properties of the segment affect the resulting allocations. An important aspect of the static allocation is its dependence on the number of providers, as the allocation has to be done accordingly. This can result in undershooting or overshooting the objective because any actual path may traverse a different number of network segments from what was assumed to be the case in the allocation scheme.
In this approach, each provider would have knowledge of how many providers are present in the traffic path and allocate among each other without wasting part of the impairment budget. Service providers may reallocate their impairment target among the segments under their control.
In this approach, providers will use signals to communicate and determine impairment budgets. In this approach, the use of resource management and signalling for the purposes of impairment apportionment is assumed. This section will consider only two kinds of signalled approach for simplicity. The first type of signalled approach is negotiated allocation approach. In this approach, networks negotiate with one another in allocating impairment budget. Starting with initial segment impairment targets, based possibly upon the static and pseudo-static allocations, the networks may negotiate for any “impairment budget” excesses, and to advertise to multiple interested parties if they can provide a network service that is within their collective impairment budget. If it is not possible to do so, the network can ask the previous network (or the user) whether more impairment budget can be allocated such that the delivery path can be determined. The second type of signalled approach is ranged allocation approach. In this approach, the range between the minimum and maximum of the allocated impairment budget for every segment along the data path is negotiated and calculated out by the use of resource management and signalling among the segments. Any value within each segment impairment budget range, when added with those of other segments, can meet the total impairment budget target for the whole data path. Thus, every segment itself can choose an appropriate value within its allocated budget range under the consideration of optimizing its resource utilization.
This approach is where possible performance levels offered by each provider are used to calculate the estimate of UNI-UNI performance and lead to decisions on path/QoS. The mechanism starts with a requesting provider determining a path that packets will follow and requesting each provider for the performance level that each will commit to. With the offers, the requesting provider will estimate the overall performance level and compare it with the UNI-to-UNI QoS class/objectives. If the path does not meet the requested objectives, the provider could take one of the following actions
As noted in section 7, there are IMT-2020 specific connectivity cases (cases #4 and #5) that deserve special treatment. Device-to-network communication is different from conventional communication in aspects such as frequency of communication (periodic) and type of traffic generated (usually more signalling traffic than data). Device-to-device communication also is distinctly different from conventional communication because the distance will be much shorter and the configuration will be simpler (with smaller number of nodes). The differences show that in-depth study is necessary to develop QoS budget allocation for these connectivity configurations. Editor’s Note: This is an important topic and we are more than welcome to have suggestions Directory: ITU-T -> focusgroups focusgroups -> International Telecommunication Union focusgroups -> Innovation of ict in Developing Countries focusgroups -> International Telecommunication Union focusgroups -> Telecommunication focusgroups -> Focus Group on Smart Grid focusgroups -> Telecommunication focusgroups -> Telecommunication focusgroups -> Telecommunication focusgroups -> Innovation of ict in Developing Countries focusgroups -> Innovation of ict in Developing Countries Download 0.91 Mb. Share with your friends:
|