Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L47/00—Traffic control in data switching networks
  • H04L47/70—Admission control; Resource allocation
  • H04L47/82—Miscellaneous aspects
  • H04L47/822—Collecting or measuring resource availability data
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L47/00—Traffic control in data switching networks
  • H04L47/10—Flow control; Congestion control
  • H04L47/20—Traffic policing
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L47/00—Traffic control in data switching networks
  • H04L47/10—Flow control; Congestion control
  • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
  • H04L47/2416—Real-time traffic
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L47/00—Traffic control in data switching networks
  • H04L47/10—Flow control; Congestion control
  • H04L47/28—Flow control; Congestion control in relation to timing considerations
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L47/00—Traffic control in data switching networks
  • H04L47/10—Flow control; Congestion control
  • H04L47/32—Flow control; Congestion control by discarding or delaying data units, e.g. packets or frames
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L47/00—Traffic control in data switching networks
  • H04L47/10—Flow control; Congestion control
  • H04L47/41—Flow control; Congestion control by acting on aggregated flows or links
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L47/00—Traffic control in data switching networks
  • H04L47/70—Admission control; Resource allocation
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L47/00—Traffic control in data switching networks
  • H04L47/70—Admission control; Resource allocation
  • H04L47/80—Actions related to the user profile or the type of traffic
  • H04L47/801—Real time traffic
• • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
  • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
  • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
  • Y02D30/00—Reducing energy consumption in communication networks
  • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Definitions

• the invention relates to a method for improving the load on a link in a communication network and a method for improving the load on a communication network.
• the total bandwidth reserved according to admission controls is higher than the bandwidth actually used.
• Networks with access controls often also use so-called policers, which measure the real bandwidth used and discard packets when the reserved bandwidth is exceeded. Bandwidth reservations are therefore usually made conservatively so that traffic peaks do not lead to the requested bandwidth being exceeded.
• connections often have an on / off character, ie there are pauses during which no data is be sent.
• data streams begin to send after a delay after the reservation is established.
• the task is for a link or a connection section with access control by claim 1 and for a
• the communication network is, for example, an IP (Internet Protocol) network.
• IP Internet Protocol
• the access control takes place either for at least one communication link in the network or for aggregated traffic transmitted over the communication network. Access control allows only a limited amount of data to use the network as a communication medium to protect it from overload.
• One possibility is also to subject only traffic of one (or more) traffic classes to an access control (for example, traffic classes for real-time traffic can be harxdeln). Traffic that is not subject to access control is then transmitted, for example, in accordance with a “best effort” approach, that is to say transmission without quality features.
• the invention allows, based on empirical values or.
• Measured values controlled allow more traffic than the pure control procedure would allow due to the displayed traffic volume.
• the overestimation of the displayed traffic descriptions i.e. the bandwidth used according to the reservation
• the basic idea of the method is to determine an empirical value or measured value for the data rate actually used by reservations in relation to the declared rate (nominal value). Measured values are included in this determination the more significant the more significant they are. For this purpose, measured values from phases with high utilization of the reservable budget are taken into account more than measured values from phases with little
• the ratios or quotients of measured values of all traffic transmitted via the link to nominal values of all traffic permitted for access control for transmission via the link are weighted according to the (absolute) size of either the measured value or the nominal value. For an overbooking based on determined relationships between measured values and nominal values, the values that relate to a traffic situation with high utilization are obviously the most significant. A high load leads to large values for the registered accumulated traffic volume (nominal value) or the measured accumulated traffic volume (measured value). The weighting takes greater account of situations with a high workload when determining an overbooking factor.
• the method according to the invention means that atypical or less frequent events are given less weight. In this way, a less conservative value can be determined for the overbooking but which takes into account the important events. This is necessary, for example, to make the determination of the overbooking factor robust against atypical situations in times of little traffic:
• the measured values are weighted in such a way that more current measurements are taken into account to a greater extent. Weighting can be done by multiplying the measured values by an aging factor. For example, all measured values are multiplied by the aging factor at fixed time intervals. Another procedure is the multiplication by an aging factor of the old measured values each time a flow is newly registered. In this case, the aging factor can be selected from the bandwidth reserved for the new flow.
• the invention can be easily combined with various types of access control schemes that utilize displayed traffic descriptions.
• the access control is carried out by means of a limit for the aggregated traffic on the link or by means of a limit for aggregated traffic transmitted via the communication network. Which aggregated traffic is measured in the communication network and compared with a nominal value for aggregated traffic depends on the access control used. Three options are outlined below:
• the admission control is carried out by means of a limit for the data traffic transmitted between two boundary points of the communication network and the measured values and the nominal values relate to the aggregated data traffic transmitted between the two boundary points.
• Two admission controls are carried out for each link affected by the transmission of data traffic by means of a link Boundary for the data traffic entering the communication network at a boundary point or with a boundary for the data traffic emerging at a boundary point. Data traffic will be allowed if all admission controls are positive. The measured values and the nominal values then relate to the aggregated permitted data traffic transmitted via an affected link, ie the aggregated data traffic transmitted via the link that entered the network at the corresponding boundary point, or to the aggregated, Data traffic transmitted via the link, which leaves the network at the other edge point.
• the method according to the invention is not limited to the above cases, but can be flexibly adapted for any access controls.
• the nominal values for aggregated traffic are compared with the limit (s); s and the measured values must correspond.
• s limit
• the aggregated traffic transmitted via the link can be measured, which has entered or is to leave the network at the corresponding edge node. Measured values should correlate with the limits. A clear correspondence is not necessary. Knowledge of the traffic distribution in the network and statistical properties of the network can be used, for example, to reduce the number of measurements required.
• the registration of a flow at the network boundary can include the following variables: The available bandwidth of links to be used for the transmission of the flow.
• a security factor which is chosen to be less than one and defines a security margin in the allocation of bandwidth.
• Such a safety factor can e.g. Intercept previous fluctuations in flows that exceed the nominal value.
• the route of the network (e.g. the route between two edge nodes) for which measured values are determined can be viewed as a virtual link with access control.
• the traffic flows between the two peripheral nodes run on physical links, which in some cases also provide bandwidth for traffic flows transmitted between other peripheral accounts.
• a virtual: link can also include alternative physical links.
• a peripheral node is also understood to be a node within the network that represents a source (or transmitter) or sink (or receiver) for data traffic. For traffic flows b> between Flows or connections that are to be transmitted via the link or between the peripheral nodes of the communication network are checked whether the transmission of the announced traffic volume (nominal value) would lead to the limit being exceeded.
• This rule applies, for example, if you set the probability value to zero.
• the reference value would be the largest ratio of measured value to nominal value that occurs.
• you can get a better utilization if you set the probability value to a small, finite value.
• the overbooking factor can then be set in proportion to the reciprocal of the reference value.
• weights are taken the weight one for measured values or nominal values above a threshold and to set the value zero for the weight below) the threshold. This means that ratios of measured values to nominal values with a low absolute value of the measured value or the nominal value are not taken into account for the determination of the overbooking factor.
• the method differentiates between different traffic classes. I.e. Measured values are determined depending on the traffic class and overbooking actuators _determined for the different traffic classes. For traffic classes with high demands on the
• Quality of service (eg so-called real-time traffic) can thus be a more conservative overbooking than for other " traffic classes " .
• Measured values for calculating an overbooking factor can be determined for the entire network and used to calculate a network-wide overbooking factor. Old natively, overbooking factors can be calculated depending on the route. For example, the access control for a new one
• Flow refer to traffic that has the same exit and exit node as the registered flow.
• measured values can be determined for the links which are used by the traffic between the input node and the output node, and an overbooking factor for this traffic can be calculated.
• the subject matter of the invention is explained in more detail below in the context of two exemplary embodiments.
• the first exemplary embodiment explains the principle of the invention on the basis of an access control relating to a link.
• the second exemplary embodiment explains which changes can be made to the first exemplary embodiment if the access control is not a link but a link
• the parameter D is the average volume of traffic that is supplied to the system as an offer during the so-called busy hour.
• the parameter D is assumed to be constant for a longer time, but it can change over hours, days and weeks.
• the variable A stores the aggregate rate of the permitted connections (i.e. all traffic transmitted via the link) according to the displayed traffic descriptions, i.e. according to the reservation request. A corresponds to the nominal value of all traffic permitted for transmission via the link. Approval is dynamic on request, which is why A (t) is a time-dependent variable.
• p ⁇ l be the targeted maximum utilization of the resource (link capacity). Since not too much traffic may be permitted on a link, A (t) ⁇ * p * B (t) applies, where ⁇ is an overbooking parameter or overbooking factor that is determined from time to time by the method according to the invention.
• the overbooking parameter is intended to compensate for the fact that the permitted connections usually do not send at their maximum rate but at a lower rate. With ⁇ > 1 the reservable capacity of the resource can be increased.
• the variable M stores the measured aggregate rate on the link, ie M is the measured value of the total traffic permitted for transmission via the link. Rates can be measured as transported data per unit of time. A moving average over a measurement window of length I M can be used to measure the aggregate rate. Like A (t), M (t) is a time-dependent " variable. There are M (t) ⁇ A (t). There are several reasons for this: o If the actual rate of an aggregate is greater than A (t), its packets are discarded by the policer at the network edge or at the node upstream of the link and its rate is therefore reduced to A (t). o Data streams begin to send after a reservation has been set up. o The traffic description of a connection is given conservatively in order to avoid losses due to rate checking in the network. o Connections have an on / off character, ie they do not always have something to send.
• P (Q ⁇ u) can be learned.
• R (t) ⁇ l applies, thus also P (Q ⁇ u) ⁇ l and thus P (Q ⁇ u) is a distribution that is continuous
• the value of U should be recalculated from time to time and ⁇ adjusted accordingly. All data from the past can be included, or a selection can be made (e.g. time window or selection of the most relevant values of Q, i.e. those for which A or M were particularly large).
• a weighting function can be used that does not take measured values from small A (t) or M (t) into account at all. This step could also be introduced in the determination of ⁇ :
• a (t) or M (t) is below a defined threshold (e.g. 0.1 * B), the measured value U is rejected and ⁇ is left at the old value instead.
• a defined threshold e.g. 0.1 * B
• the duration of a measurement interval I M should be of the order of magnitude below a connection duration ( ⁇ 10s), since otherwise some connections cannot be recorded at all.
• the duration of a measuring circuit should be chosen so large that sufficient statistical data are available to estimate U well.
• the method can also be given more or less memory by keeping the statistics for Q (t) over a more or less long time (several measurement cycles).
• This procedure can be transferred from the access control of a pipe or a link to the access control of a network.
• the figure shows a communication network with access control.
• Edge nodes are indicated by filled circles, inner nodes by empty circles. On the left are represented by connections between the nodes.
• An input node is labeled I, an output node E, and a link L, for example. Part of the traffic between nodes I and E is transmitted via link L.
• An access control at the input node I and at the output node E together with access controls at other edge nodes ensure that no overload occurs at the link L.
• the access control is carried out by means of a limit or a budget B (I, E) for the traffic transmitted between the nodes I and E.
• the method described for a link can be used for the communication network by extending the variables to border-border (b2b) relationships or edge-to-edge relationships between the access node I and the output node E.
• any such relationship is considered a virtual link.
• o D becomes D (I, J)
• o B becomes B (I, J)
• A becomes A (I, J)
• o M becomes M (I, J).
• M (v, w) can be determined on the policer, for example. But if that's through If measurements are to be made on links, one can measure their overall rate and divide this proportionally over all active border-b2b relationships proportional to A (I, J). The maximum of these measured values could then be taken for M (I, J), for example.
• o Q becomes Q (I, J).
• o U becomes U (I, J).

Landscapes

• Engineering & Computer Science (AREA)
• Computer Networks & Wireless Communication (AREA)
• Signal Processing (AREA)
• Data Exchanges In Wide-Area Networks (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=34305515

Family Applications (1)

Country Status (5)

Families Citing this family (5)

Family Cites Families (6)

• 2003
  • 2003-09-05 DE DE10319310A patent/DE10319310B4/de not_active Expired - Fee Related
  • 2003-11-21 EP EP03788811A patent/EP1661329A1/de not_active Withdrawn
  • 2003-11-21 US US10/570,430 patent/US20060274653A1/en not_active Abandoned
  • 2003-11-21 WO PCT/DE2003/003873 patent/WO2005027432A1/de not_active Application Discontinuation
  • 2003-11-21 CN CNA200380110661XA patent/CN1879358A/zh active Pending

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events