HK1089888A - System and method for providing differentiated services - Google Patents

Description

System and method for providing differentiated services

Technical Field

The present invention relates to communication systems and methods, and more particularly to systems and methods for providing differentiated services to telecommunications traffic flows in a resource efficient manner.

Background

The internet is a packet-based network that transports many different types of telecommunications traffic (e.g., voice, data, and multimedia traffic) that originate from a variety of applications. Different types of traffic have different quality of service (QoS) requirements. For voice traffic, small and uniform packet delay is particularly important, while for data traffic, less packet loss is the most important requirement. Furthermore, service providers are interested in providing services with different QoS that allow them to meet the varying needs of their subscribers and maintain a differentiated pricing scheme. Several mechanisms have therefore been developed for providing different QoS to different users and traffic flows in packet-based networks (e.g., the internet).

International patent application WO02/25867 describes a radio access network that provides different priorities for different packet data connections with user equipment. The priority of the data connection may be dynamically adjusted by the control node according to throughput criteria communicated to the control node by the user equipment.

The IETF (internet engineering task force) developed an integrated services (IntServ) architecture described in IETF RFC 1633. The IntServ architecture uses an explicit mechanism to signal per-flow QoS requirements to network elements such as hosts and routers. There are a number of disadvantages associated with IntServ. IntServ requires maintenance and control of per-flow state and classification. Reserving network resources on a per flow basis introduces scalability issues for the core network, where the number of flows processed is often in the million range. The IntServ architecture can only be practically used for small access networks with a moderate number of streams.

To overcome the scalability and complexity problems of IntServ, IETF introduced the differentiated services (DiffServ) architecture described in IETF rfc 2475. Traffic passing through a network core router implementing DiffServ is processed on an aggregated basis. Traffic entering the network is classified and assigned to different behavior aggregates. Each behavior aggregate is identified by a single DS (differentiated services) codepoint. When classifying traffic, packets are marked with a specific DS code point, which is placed in the DS field in the IP (internet protocol) header. Within the core of the network, the data packet is forwarded according to a single hop behavior (PHB) associated with the DS code point of the data packet. The PHB determines externally observable forwarding behavior (e.g., forwarding delay and packet loss) for nodes at different load levels. A PHB is a logical network resource that controls the use of underlying physical network resources. A PHB may therefore be considered a partial network resource that defines a subset of the total network resources.

International patent application WO02/11461 is an example of a document describing a differentiated services system. It discloses methods and apparatus for providing dynamic quality of service through bandwidth brokers in IP networks including the DiffServ architecture. The bandwidth broker may obtain resource availability information by communicating only with the edge routers of the DiffServ domain.

Another example of a document discussing the implementation of DiffServ is international patent application WO02/080013, which describes dynamic resource allocation for providing differentiated services over a broadband communication network comprising satellites.

A service is typically specified when a service provider sells a bearer service to an end user. A service may be specified by a Service Level Specification (SLS) that includes QoS requirements for the service. SLSs may therefore be used to define different classes or classes of service. In order to meet the QoS requirements of SLSs, the allocation of resources allocated to the traffic associated with the SLSs is very important. Currently, the mapping of SLSs to network resources is typically done semi-permanently as part of network provisioning and configuration, see IETF RFC 3086. This mapping is set to meet the expected traffic mix.

In conventional networks, the traffic characteristics are completely well known. In future multi-service networks and multi-access networks, traffic characteristics will be dynamic due to changes in user behavior, introduction of new applications, etc. Moreover, these variations mean that the network must be flexible in terms of resource allocation for different QoS classes. New techniques for efficiently managing networks are therefore needed.

Summary of The Invention

As mentioned above, current networks are sized according to some expected traffic mix. The network elements and mechanisms are statically configured through the management interface. If the traffic mix changes, a lot of work is required to reconfigure the network. This is often more costly than having the network work with a sub-optimal configuration. In networks with very dynamic traffic mixes, it is therefore possible to improve the network resource utilization considerably if the mapping of service classes to network resources can be changed relatively easily on a time scale much shorter than what is currently common.

It is therefore an object of the present invention to provide an apparatus and method that allows dynamic mapping of service classes to network resources, such that efficient resource utilization can be achieved even when the traffic mix changes.

The above object is achieved by a system according to claim 1, a method according to claim 11 and a control device according to claim 20.

The arrangement and method according to the invention make it possible to dynamically and automatically change the mapping of traffic to partial resources based on information about the actual traffic mix currently being transmitted in the network. The mapping is adapted to the traffic mix to obtain a mapping that enables a more efficient use of all resources when forwarding the traffic mix while satisfying the set service requirements. The optimal mapping is generally considered to be the mapping that minimizes the amount of wasted resources, i.e. resources that are not used for forwarding the traffic mix, but are reserved in a way that they cannot be used for transmitting other traffic.

According to a first aspect of the present invention, a system for forwarding telecommunication traffic of a plurality of microflows in a quality of service enabled telecommunication network is provided. Each microflow is assigned a service level according to a set of predetermined service levels to establish a set of service level aggregate flows, and each service level is associated with a set of service requirements. The system comprises a set of partial resources for forwarding traffic in the network, to which partial resources service level aggregate flows are mapped. The system further comprises control means arranged to receive information on traffic characteristics and resource performance of each service level aggregate flow and to update the mapping of the service level aggregate flows to the set of partial resources based on the received information, to obtain an updated mapping that reduces the total amount of wasted resources while fulfilling the service requirements of the service levels.

According to a second aspect of the present invention, a method is provided for updating a mapping of service level aggregate flows to a set of partial resources for forwarding traffic in a quality of service enabled telecommunications network. The method includes the step of receiving a set of service level aggregate flows. Each service level aggregate flow is made up of microflows that are assigned the same service level from a set of predetermined service levels, and each service level is associated with a set of service requirements. The method further comprises the steps of obtaining information about traffic characteristics and resource performance of each service level aggregate flow, and updating the mapping of the service level aggregate flows to the set of partial resources based on the obtained information to obtain an updated mapping that reduces the total amount of wasted resources while satisfying the service requirements of the service levels.

According to a third aspect of the present invention, control means are provided for controlling a mapping of service level aggregate flows of telecommunications traffic to a set of partial resources for forwarding the traffic in the network. Each service level aggregate flow corresponds to a service level associated with a set of service requirements. The control means comprises means for receiving information about traffic characteristics of each service level aggregate flow and performance of the set of partial resources. Furthermore, the control means comprises means for dynamically controlling the mapping of the service level aggregate flows to the set of partial resources based on the received information to obtain an updated mapping that reduces the total amount of wasted resources while fulfilling the service requirements of the service levels.

According to a preferred embodiment of the invention, the resource parameters controlling the performance of the partial resources are updated in view of the current traffic mix in order to achieve a combination of resource parameters and mapping that yields the most efficient utilization of the partial resources, which is a mapping that minimizes the total amount of wasted resources and also meets the service requirements of the service classes.

In order to optimize the utilization of network resources with several service classes, the mapping of different service classes to available network resources must be adjusted in a dynamic manner to adapt to the current traffic mix. In contrast to the semi-fixed configuration of such mapping, which is implemented in prior art products, the present invention enables the network operator to perform adaptive mapping in a dynamic manner.

Mapping, which is performed semi-fixedly as in prior art solutions, may become inefficient when the traffic mix changes. The fact that the mapping is inefficient may not be detected and even if it is detected, it is often cumbersome to change the mapping that is performed semi-fixedly.

In contrast, the present invention has the advantage that it allows to continuously monitor the current traffic mix and adapt the mapping accordingly. According to the invention, the mapping can be adjusted quickly and automatically when the traffic mix changes.

The invention makes it possible for network operators to utilize their network resources more efficiently. Thus, the operator may be able to forward more traffic, provide better quality of service or reduce the amount of network resources.

Since the invention allows to continuously monitor the current traffic mix and dynamically optimize the mapping of service classes to network resources, the invention also enables the network operator to more easily determine that the service requirements of different service classes are fulfilled and that no network resources are overloaded without having to provide an excess capacity of network resources.

Further advantages and objects of embodiments of the present invention will become apparent when reading the following detailed description in conjunction with the drawings.

Brief Description of Drawings

FIG. 1 is a schematic block diagram illustrating a system in which the present invention may be used.

Fig. 2 is a schematic block diagram illustrating an embodiment of a mechanism for dynamically mapping between aggregate flows and partial resources in accordance with the present invention.

FIG. 3 is a schematic block diagram illustrating a mechanism for dynamic mapping between aggregate flows and single hop behaviors (PHBs) in a network using a differentiated services architecture in accordance with the present invention.

Fig. 4 is a schematic block diagram illustrating an alternative embodiment of the mechanism illustrated in fig. 3.

Fig. 5 is a diagram illustrating a mapping of service level aggregate flows to PHBs.

Fig. 6 is a schematic diagram illustrating how the mapping shown in fig. 5 may be changed in order to provide more efficient utilization of resources in accordance with the present invention.

Detailed Description

The present invention now will be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the scope of the invention to those skilled in the art. In the drawings, like numbering represents like elements.

According to the invention, the mapping of service classes to resources is dynamically and automatically changed based on feedback or signaling information giving information about the actual traffic mix currently being transmitted in the network. For example, the present invention may be used to map UMTS bearer services to DiffServ PHBs.

The present invention is based on dynamic optimization of the mapping and is generally illustrated by the diagram of fig. 1. Fig. 1 illustrates multiple packet streams being multiplexed on the same physical link 31. These streams may be generated by, for example, telephony, video conferencing, streaming, and interactive applications. Therefore, these streams have different requirements on bandwidth, transmission delay, and packet loss rate. It is therefore advantageous to process flows according to different service classes and classify the traffic accordingly. Flows with the same or similar QoS requirements are assigned the same class of service and treated as one aggregate traffic flow to reduce complexity.

It is assumed that each packet flow has explicit requirements on bandwidth, delay and packet loss. In fig. 1, a1 indicates a telephone application, a2 indicates a video conference application, and the like. The traffic mix to be transported over the physical link 31 consists of n1 flows from application a1, n2 flows from application a2, etc. Flows from application a1 are assigned a first service class with QoS requirements R1 and form a service class aggregate flow S1, similarly flows from application a2 are assigned a second service class with QoS requirements R2 and form a service class aggregate flow S2, and so on. Note, however, that flows from different applications may be combined into a single aggregate flow, if this is appropriate in view of the QoS requirements of the flows.

To control the traffic entering the physical link, a plurality of buffers B1, B2. Each buffer is allocated a portion of the available bandwidth on the physical link. Thus, the buffer represents a set of partial resources on which traffic is to be distributed. The mapping rules determine a mapping between the aggregate flows S1, S2.. SM and the partial resources B1, B2.. BN. Furthermore, a given multiplexing rule determines the arrangement of the transmission of data packets belonging to the respective partial resource on the link 31. The scheduling determines that the delay and loss targets for each stream are met.

In order to achieve as efficient a utilization of resources as possible, it is desirable to ensure that the traffic transmission capabilities of the partial resources can be used to the maximum extent. This means that traffic flows should be fed to partial resources so that as much traffic as possible can be transmitted through the partial resources. If the traffic transmission capability of a partial resource is not maximally utilized in consideration of the current traffic mix, a part of the partial resource that is not used for transmitting traffic is a wasted resource. The term "wasted resources" is defined in this specification as resources that are not used for transmitting a given mix of services, but are still reserved such that they cannot be used for transmitting other services. The implications of wasted resources will be further explained below in conjunction with the figures.

The optimal mapping can be defined as follows:

for a given traffic mix consisting of aggregate flows s1.. SM and for a given set of partial resources b1.. BN, the optimal mapping minimizes wasted resources when transmitting the traffic mix in order to meet the requirements r1.. RM.

The network operator may choose an alternative definition of the best mapping, but typically the operator is interested in minimizing the amount of wasted resources that occur when a given amount of traffic is transmitted while meeting the QoS requirements associated with the traffic. The operator may be able to make certain resources available for transmitting additional traffic.

If it is possible to change parameters associated with and affecting the performance of the partial resources, a better utilization of the partial resources may be obtained by determining a combination of mapping and resource parameters that minimizes wasted resources while satisfying the service level QoS requirements. The resource parameters controlling the performance of the partial resources may be, for example, parameters such as buffer size and priority assigned to the different partial resources. These parameters can affect the underlying physical resource portion allocated to the partial resource. The choice of the mechanism for arranging the partial resources to access the physical link also affects the performance of the partial resources. The resource parameters may affect the performance of the partial resources in a manner that affects packet delay and packet loss for the partial resources at different loading levels. The ability to adjust the fractional resources makes it possible to minimize the total amount of network resources allocated for the transmission of a given set of aggregated traffic streams.

As stated above, there is a direct translation between the mapping and scheduling rules and the cost of transporting the traffic mix. The optimal mapping is thus the mapping that minimizes the bandwidth cost.

Different service mixes may have different optimal mapping and scheduling rules. This means that when the traffic mix on a given link changes, the operator can reduce costs by dynamically changing the mapping or by dynamically changing both the mapping and the scheduling rules.

Obviously, the above definition of the best mapping as the one that minimizes the bandwidth cost corresponds to the definition of the best mapping as the one that maximizes the operator's revenue.

FIG. 2 is a block diagram illustrating an embodiment of a mechanism for performing dynamic optimization of mapping and scheduling rules in accordance with the present invention. Fig. 2 shows L micro flows f1, f2, fL arriving at a network node and requesting access to a physical network element 21, such as a link. Microflow is a single example of an application to application flow. Microflows originate from different applications and have different requirements on QoS such as bandwidth, delay and packet loss. These requirements are signaled to and negotiated with the entry control function 22. The entry control function assigns a Service Level Specification (SLS) to a particular microflow whenever link resources are available. The service level specification includes a traffic throttling specification (TCS) that specifies traffic characteristics such as peak rate, average rate, maximum allowable burstiness that the microflow must meet when entering the network. The SLSs define different classes or classes of service and the microflows are allocated SLSs corresponding to their respective QoS requirements. Flows having the same or similar QoS requirements are assigned the same SLS, thereby forming a plurality of aggregate traffic flows. Fig. 2 shows M aggregate flows SLS 1. Traffic parameters such as average bandwidth, burstiness are measured separately for each aggregate flow by the measurement function 26. According to the present invention, these parameters are reported as feedback information to the mapping control unit 23. The feedback information makes the mapping control unit 23 aware of the characteristics of the actual traffic mix, e.g. the ratio between the traffic of each aggregate flow and the traffic of the different aggregate flows.

The mapping control means 23 is responsible for programming the mapping function 24 based on the received feedback information. The mapping function 24 maps the aggregate flow to N partial resources B1. The QoS level of a partial resource is determined by a multiplexing rule or scheduling rule that controls how the partial resource is multiplexed onto the physical resource. Different portions of the total underlying physical resource may be allowed to be used by different partial resources. This may be controlled by, for example, a round robin scheduling mechanism or other mechanism as is well known to those skilled in the art.

In the embodiment shown in fig. 2, the multiplexing of part of the resources is controlled by a multiplexing function 25, which multiplexing function 25 can be reprogrammed under the control of the mapping control means 23. The capacity of the partial resources can thus be changed in response to feedback information received by the mapping control means about the service mix. The mapping control unit may also receive feedback information from the measurement function 27 regarding delay and loss data per resource packet portion. The mapping control unit may thus detect whether a part of the resources is or will become overloaded, which further assists the mapping control unit in determining the optimal mapping and multiplexing rules. The feedback information from the measurement function 27 may also give an indication as to whether the QoS requirements of each microflow are fulfilled or not. The delay and packet loss performance of the microflows on the network element can also be measured end-to-end and the results can be compared with the delay and loss rate requirements of the flows specified for the microflows in the SLS.

Using the above-mentioned feedback information on traffic characteristics, resource load and comparison of actual QoS and QoS requirements, the mapping control means determines the optimal mapping and scheduling rules by means of an optimization algorithm. As mentioned above, the optimal mapping and scheduling rules are generally considered to be mapping and scheduling rules that minimize the total amount of wasted resources and thus also the utilization of the network element 21. Feedback information from the measurement functions 26 and 27 to the mapping control unit allows the system to adapt in real time to changes in the traffic mix.

In the embodiment illustrated in fig. 2, the characteristics of the partial resource B1. Thus, the mapping control means is able to influence the resource utilization by controlling the scheduling rules, i.e. the resource characteristics, and by controlling the mapping of aggregate flows to partial resources. Even if the scheduling rules are fixed such that the characteristics of the partial resources cannot be changed, the mapping can be set to a mapping that is optimal in view of the available partial resources. However, in more flexible systems where it is possible to adjust the partial resources, it is generally possible to obtain a more efficient total resource utilization than in less flexible systems with fixed partial resources.

Fig. 3 illustrates an embodiment of the present invention in an IP QoS network using a differentiated services QoS architecture. In this architecture, the portion of resources to which aggregate flows are mapped is referred to as single hop behavior (PHB). In fig. 3, the mapping of aggregate flow SLS1, SLSM to phb1, PHBN is illustrated. A PHB is an allocated buffer and link bandwidth resource that determines the externally observable forwarding behavior (e.g., forwarding delay or packet loss) of a node. In the embodiment shown in fig. 3, the mapping control means controls the mapping of aggregate flows to PHBs in response to received feedback information from the measurement functions 26 and 27. The PHBs are arranged by the arranging function 28 onto the link 31. The scheduling function 28 is programmed by the mapping control device so that the PHBs can be optimized according to the currently received service mix.

In the embodiment of the invention shown in fig. 2 and 3, the aggregate flows SLS1, the traffic parameters of the SLSM are measured by a measurement function and reported to the mapping control. The mapping control means is thus provided with information about the traffic characteristics of the aggregate flows, which information is used to determine an optimal mapping in view of the traffic mix. According to an alternative embodiment of the invention, the information on the traffic characteristics reported to the mapping control unit is based on calculations rather than measurements. During the setup of microflows, it may be determined that a certain microflow may not exceed certain traffic limits, for example in terms of average rate and peak rate. These traffic restrictions may be reported to the admission control function 22 by RSVP or ATM signaling. The admission control function 22 may then calculate the traffic limit for each aggregate flow based on the traffic limits of the microflows included in the respective aggregate flow. The calculated traffic limit for each aggregate flow may then be reported from the admission control function 22 to the mapping control means 23 as information about the traffic characteristics of the aggregate flow. According to this alternative embodiment of the invention, the measurement function 26 may therefore be omitted, as shown in fig. 4. Alternatively, the measurement function 26 may be arranged such that it is capable of receiving the traffic limitation calculation results from the admission control device and may be arranged to report the measurement results or the received traffic limitation calculation results to the mapping control device 23.

If the information reported to the mapping control means about the traffic characteristics of the aggregate flows is based on the calculated traffic limit, it is possible to adjust the mapping to accommodate a slightly overestimated traffic volume, since the microflows may be below the established traffic limit rather than above it. It is thus possible that a mapping based on measured information about the traffic mix is generally more resource-saving than a mapping based on calculated information.

To further illustrate the functionality of the system for controlling partial resource utilization according to the present invention, a specific simple example of an optimization algorithm that may be used by the mapping control unit is described below with reference to fig. 5 and 6.

Fig. 5 illustrates the optimization principle for an example of mapping seven service level aggregate flows SLS1, SLS2, SLS7 to three PHBs PHB1, PHB2, PHB 3. Each PHB has two parameters: the peak rate and the average rate are allocated to its underlying physical resources. Also, each service level aggregate flow is associated with requirements for peak rate and average rate. In fig. 5, the peak rate is indicated on the x-axis, and the ratio between the average rate and the peak rate is indicated on the y-axis. The average rate is thus the area of the PHB box or aggregate flow box. In this example, all service level aggregate flows mapped to a particular PHB have the same peak rate burst time scale, and this time scale matches the PHB's buffer size.

If the ratio of the average rate to the peak rate is 1, then PHB resources are allocated to transmit the peak rate of the traffic. At this point no packet loss or queue delay occurs. If the ratio is less than 1, the operator sells more peak rate service level specifications than the network can handle immediately. Traffic must be buffered. As a result, queuing delays, and even packet losses, may occur.

The diagram in fig. 5 gives an illustration of the optimization problem, such as the problem of packing aggregate flow boxes SLS1, SLS7 into PHB boxes PHB1, PHB2, PHB3 in the most efficient way. When the aggregate flow box SLS 1.. the SLS7 stays within the limits of the PHB box, the underlying physical resources of the PHB may support the requirements of the aggregate flow in terms of peak and average rates. The area of a PHB box not covered by an aggregate flow box indicates wasted average rate resources.

The optimization algorithm according to an embodiment of the present invention minimizes the waste of average rate resources by moving the boundaries between PHB boxes along the x-axis and by adjusting the height of the PHB box so that it equals the highest aggregate flow height within the box. This is shown in FIG. 6, where the boundary between the PHB1 and PHB2 boxes has been moved to the left, and the heights of the two boxes have been adjusted to the highest aggregate flow within each box. The adjusted PHB boxes are labeled PHB1+ and PHB2+ in fig. 6.

It can be seen that within the PHB1+ box compared to PHB1, wasted resources have been reduced. On the other hand, there is a slight increase in wasted resources within the PHB2+ box compared to PHB 2. However, the reduced waste is higher than the increased waste, resulting in a net saving of resources.

The optimization algorithm involves the mapping function 24 and the scheduling function 28 of fig. 3, as described below. Moving the boundary of the PHB frame along the x-axis means that the mapping of aggregate flows onto the PHB must be changed. This is achieved by reprogramming the mapping function 24. Also, the PHB's buffering and bandwidth resources are adjusted as the PHB frame's boundaries are moved along the x or y axis. This is accomplished by reprogramming the scheduling function 28.

For the special case with two PHB boxes, the aggregate flow boxes have equal width and their height represents the sequence of geometric reductions { a }_iThe following optimization algorithm may be used to recursively update the mapping of service level aggregate flows to two PHB boxes:

(a₁-a_T+1)/(a_T+1-a_T+2) ＜(M-T-1) ＝＞T：＝T+1

(a₁-a_T)/(a_T-a_T+1) ＞(M-T) ＝＞T：＝T-1

m is the total number of service level aggregate flows and T is the number of service level aggregate flows in the first of the two PHB boxes. The algorithm simply states T: th service level aggregate flow, SLST, should be moved from a first PHB to a second PHB, or T + 1: th service level aggregate flow, SLST +1, should be moved from the second PHB to the first PHB.

After moving the service level aggregate flows, the resources of the PHB must be adjusted accordingly. This adjustment may be calculated based on a priori knowledge of the resource requirements of the service level aggregate flows, or based on PHB performance measurements in terms of delay and packet loss.

The two-dimensional optimization algorithm outlined herein can be generalized to multi-dimensional cases that include additional parameters describing the traffic and resources, such as various leaky bucket parameters. However, it is possible that a good recursive algorithm to find the global optimum does not exist for the general case.

The most straightforward forwarding method is to compute the amount of wasted resources for all possible mappings of service level aggregate flows to PHBs and choose the best. To reduce the number of combinations, service level aggregate flows should be grouped according to their similarity in terms of resource requirements, as is the case in fig. 5.

In the above algorithm, each service level aggregate flow is mapped onto a single PHB. However, according to alternative embodiments of the present invention, the service level aggregate flows may be split between two or more resources, such as PHBs. In some cases, splitting the service level aggregate flow among several resources may result in more efficient utilization of the resources. This may be particularly advantageous in situations where it is for some reason not allowed or possible to adjust the resource characteristics.

In case the traffic mix changes often, the above algorithm may have the effect of changing the mapping very often. Moving service level aggregate flows back and forth between different PHBs multiple times during a short period of time may have a negative impact on network performance. To eliminate this negative effect, an algorithm that includes some type of hysteresis may be used. For example, a service level aggregate flow may be moved from one PHB to another only when the reduction in wasted resources exceeds a certain limit, or a minimum time period may be specified between two successive readjustments of the mapping of the service level aggregate flows to resources.

A change in traffic mix will result in a change in service level aggregate flow characteristics. In fig. 5 and 6 this would have the effect of changing the area of the SLS7, block SLS 1. After such a change, it may be advantageous to rearrange the mapping of aggregate flows to partial resources in order to achieve a more efficient use of the total resources. Since the mapping control apparatus according to the present invention receives information on the currently received service mix, the present invention makes it possible to quickly detect and adapt to a change in the service mix. The information from the measurement function 26 or the admission control function 22 may include information about traffic characteristics such as average rate, peak rate, and/or some other characteristic of the service level aggregate flows.

The information about the current traffic mix measured or calculated and reported to the mapping control unit according to the invention can also be used for other purposes than optimizing resource utilization. It may also be used to determine parameters used to optimize the performance of the QoS mechanisms implemented in the routers. There are implementations such as the DiffServ architecture where it is important to configure the desired packet average length. This information can of course be used to change the DiffServ configuration parameters if significant changes in average packet length can be detected when traffic is measured in the measurement function 26.

The mapping control means is the central unit of the invention. It is responsible for determining that a portion of the resources are used in an efficient manner according to the currently received traffic mix without overloading. As will be apparent to a person skilled in the art, several different implementations of the mapping control apparatus are possible. It is possible that e.g. each node of a network or QoS domain is provided with mapping control means or that the mapping control means are provided in a central control node communicating with the network nodes. It will be apparent to those skilled in the art how known hardware and software components may be used to implement the mapping control apparatus and other functions of the present invention. The mapping function 24 is according to the invention implemented to be programmable under the control of a mapping control device. The easiest way to implement the programmable mapping function is by software, but programmable hardware implementations and combined hardware and software implementations are also possible. As mentioned above, a preferred feature of the invention is that the multiplexing function 25 or scheduling function 28 is also programmable under the control of the mapping control means.

In the drawings and specification, there have been disclosed typical preferred embodiments of the invention and, although specific terms are employed, they are used in a generic and descriptive sense only and not for purposes of limitation, the scope of the invention being set forth in the following claims.

Claims (28)

1. A system for forwarding telecommunication traffic of a plurality of micro flows (f1, fL) in a telecommunication network allowing quality of service, each micro flow being assigned a service level from a set of predetermined service levels to produce a set of service level aggregate flows (S1, S2, SM, SLS1, SLSM), wherein each service level is associated with a set of service requirements (R1, R2, RM), the system comprising a set of partial resources (B1, BN, PHB1, PHBN) for forwarding traffic in the network, to which partial resources the service level aggregate flows are mapped, characterized in that the system further comprises control means (23), said control means (23) being arranged to

Receiving information about the service characteristics and resource performance of each service level aggregate flow; and

the mapping of the service level aggregate flow to the set of partial resources is updated based on the received information to obtain an updated mapping that reduces the total amount of wasted resources while meeting the service requirements of the service level.

2. A system according to claim 1, characterized in that said control means (23) is further arranged to update resource parameters controlling the performance of the set of partial resources to achieve a combination of resource parameters and mapping that minimizes the total amount of wasted resources while fulfilling service requirements (R1, R2, RM) of service classes.

3. The system of claim 2, wherein the resource parameter is a scheduling parameter controlling how the set of partial resources is multiplexed onto the physical link (31).

4. The system according to any of the preceding claims, characterized in that the set of partial resources is a set of DiffServ single hop behaviors (PHB1, PHBN).

5. The system according to any of the preceding claims, further comprising a resource performance measurement function (27) for measuring resource performance and reporting the measurement results to a control device (23), and said information on resource performance is information on packet delay and packet loss for each partial resource (B1, BN, PHB1, PHBN) measured by said resource performance measurement function.

6. The system according to any of the preceding claims, further comprising a traffic measurement function (26) for measuring traffic characteristics of at least one service level aggregate flow and reporting the measurement results to the control means (23), and wherein at least part of said information on traffic characteristics of individual service level aggregate flows is information on average and peak rates measured by said traffic measurement function.

7. A system according to any of the preceding claims, characterized in that the system further comprises means (22) for receiving signalling information about traffic restrictions set up for the micro flows (f1, fL), means for calculating a respective traffic restriction for each service level aggregate flow based on the signalling information about each micro flow traffic restriction, and means for sending the traffic restriction for each service level aggregate flow to the control device (23) as at least part of the information about the traffic characteristics of the service level aggregate flows (S1, S2, SM, SLS1, SLSM).

8. A system according to any one of the preceding claims, characterized in that the control means (23) is arranged to map each service level aggregate flow (S1, S2, SM, SLS1, SLSM) to one or several partial resources (B1, B2, BN, PHB1, PHBN) in the set of partial resources.

9. A system according to any one of the preceding claims, characterized in that said control means (23) is arranged to use a recursive optimization algorithm to minimize the total amount of wasted resources taking into account the received information and the set of service requirements (R1, R2, RM).

10. A system according to any preceding claim, wherein the control means (23) is arranged to update the mapping when the received information indicates that at least one traffic characteristic of at least one service level aggregate flow (S1, S2, SM, SLS1, SLSM) has changed by a predetermined amount since the mapping was last updated.

11. A method for updating a mapping of service level aggregate flows to a set of partial resources (B1, B2, BN, PHB1, PHBN) for forwarding traffic in a quality of service enabled telecommunications network, comprising the steps of:

receiving a set of service level aggregate flows (S1, S2, SM, SLS1, SLSM), wherein each service level aggregate flow is constituted by a micro-flow, said micro-flow having been allocated the same service level from a set of predetermined service levels, each service level being associated with a set of service requirements (R1, R2, RM), characterized by the further steps of

Obtaining information about the service characteristics and resource performance of the service level aggregate flows; and

updating the mapping of the service level aggregate flow to the set of partial resources based on the obtained information to obtain an updated mapping that reduces the total amount of wasted resources while meeting the service requirements of the service level.

12. The method of claim 11, wherein the step of updating the mapping comprises updating resource parameters that control performance of the set of partial resources to achieve a combination of mapping and resource parameters that minimizes a total amount of wasted resources while satisfying service requirements (R1, R2, RM) for a service level.

13. The method of claim 12, wherein the resource parameter is a scheduling parameter controlling how the set of partial resources is multiplexed onto the physical link (31).

14. The method according to any of the claims 11-13, characterized in that the step of obtaining information comprises measuring packet delay and packet loss for each partial resource (B1, B2, BN, PHB1, PHBN).

15. The method of any of claims 11-14, wherein the step of obtaining information comprises measuring an average rate and a peak rate of at least one service level aggregate flow (S1, S2, SM, SLS1, SLSM).

16. The method according to any of the claims 11-15, wherein the step of obtaining the information on the traffic characteristics of the service level aggregate flows (S1, S2, SM, SLS1, SLSM) comprises the steps of receiving signaling information on the traffic limits set up for the micro flows (f1, fL), and calculating the respective traffic limit for each service level aggregate flow based on the signaling information on each micro flow traffic limit to form at least a part of the information on the traffic characteristics of the service level aggregate flows.

17. The method according to any of claims 11-16, characterized by updating the mapping such that each service level aggregate flow (S1, S2, SM, SLS1, SLSM) is mapped to one or several partial resources (B1, B2, BN, PHB1, PHBN) in the set of partial resources.

18. The method according to any of the claims 11-17, characterized in that the updating of the mapping is performed using a recursive optimization algorithm in view of the obtained information and the set of service requirements (R1, R2, RM) in order to minimize the total amount of wasted resources.

19. The method according to any of the claims 11-18, wherein the updating of the mapping is performed when the obtained information indicates that at least one traffic characteristic of at least one service level aggregate flow (S1, S2, SM, SLS1, SLSM) has changed a predetermined amount since the last updating of the mapping.

20. Control device (23) for controlling the mapping of service level aggregate flows (S1, S2, SM, SLS1, SLSM) of telecommunication traffic to a set of partial resources (B1, B2, BN, PHB1, PHBN) for forwarding traffic in a network, wherein each service level aggregate flow corresponds to a service level associated with a set of service requirements (R1, R2, RM), characterized in that said control device (23) comprises

Means for receiving information about traffic characteristics of each service level aggregate flow and performance of the set of partial resources; and

means for dynamically controlling a mapping of a service level aggregate flow to the set of partial resources based on the received information to obtain an updated mapping that reduces a total amount of wasted resources while satisfying service requirements of the service level.

21. The control apparatus of claim 20, wherein the control apparatus further comprises means for dynamically controlling resource parameters that govern the performance of the set of partial resources to achieve a combination of mapping and resource parameters that minimize a total amount of wasted resources while meeting service requirements of a service level.

22. The control apparatus according to claim 21, wherein said resource parameter is a scheduling parameter controlling how the set of partial resources is multiplexed onto the physical link (31).

23. A control arrangement according to any of claims 20-22, characterized in that the control arrangement further comprises means for sending information about the current mapping to an admission control arrangement (22), which admission control arrangement (22) controls the admission of micro flows (f1, fL) into service level aggregate flows (S1, S2, SM, SLS1, SLSM).

24. The control apparatus according to any of claims 20-23, wherein the information on the performance of the set of partial resources comprises information on packet delay and packet loss of the respective partial resource.

25. The control apparatus according to any of claims 20-24, wherein the information on traffic characteristics of each service level aggregate flow comprises information on average rate and peak rate.

26. A control arrangement according to any of claims 20-25, c h a r a c t e r i z e d i n that said means for dynamically controlling the mapping is arranged to control the mapping such that each service level aggregate flow (S1, S2, SM, SLS1, SLSM) is mapped to one or several partial resources (B1, B2, BN, PHB1, PHBN) in the set of partial resources.

27. A control apparatus according to any one of claims 20 to 26, wherein the means for dynamically controlling the mapping comprises a computing means arranged to use a recursive optimisation algorithm to minimise the total amount of wasted resources taking into account the received information and the set of service requirements.

28. A control arrangement according to any of claims 20-27, c h a r a c t e r i z e d i n that the control arrangement is arranged to update the mapping when the received information indicates that at least one traffic characteristic of at least one service level aggregate flow (S1, S2, SM, SLS1, SLSM) has changed by a predetermined amount since the last update of the mapping.