US20140160945A1 - Using bandwidth measurements to adjust cir and eir on a sub-rate link - Google Patents
Using bandwidth measurements to adjust cir and eir on a sub-rate link Download PDFInfo
- Publication number
- US20140160945A1 US20140160945A1 US13/706,420 US201213706420A US2014160945A1 US 20140160945 A1 US20140160945 A1 US 20140160945A1 US 201213706420 A US201213706420 A US 201213706420A US 2014160945 A1 US2014160945 A1 US 2014160945A1
- Authority
- US
- United States
- Prior art keywords
- synthetic
- packet
- location
- packets
- network segment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/28—Flow control; Congestion control in relation to timing considerations
- H04L47/283—Flow control; Congestion control in relation to timing considerations in response to processing delays, e.g. caused by jitter or round trip time [RTT]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0852—Delays
- H04L43/087—Jitter
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0894—Packet rate
-
- 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
- H04L47/22—Traffic shaping
Definitions
- This invention is directed towards the ability to react to a detected change in the available bandwidth of a sub-rate link, for instance wireless, by adjusting operating parameters for the link with regard to service assurance, traffic engineering, service reliability and related functions.
- Time synchronization protocols such as IEEE 1588v2 also require a link with a stable bandwidth in order to derive and maintain clock synchronization between a pair of networking devices.
- Flow control mechanisms have been used to adapt to the bandwidth of a link or network segment (ref: U.S. Pat. No. 8,139,485). Such flow control protocols require the knowledge of the protocol at each node in the segment.
- a method and system for measuring the bandwidth of a network segment between a first location and a second location by generating a plurality of synthetic packets at the first location and marking each generated synthetic packet with a transmission timestamp indicative of a first clock time at which the synthetic packet is generated; transmitting each of the generated synthetic packets to the second location; receiving each synthetic packet at the second location and marking each received synthetic packet with a reception timestamp indicative of a second clock time at which the synthetic packet is received at the second location; calculating an inter-packet delay variation of the network segment based on the transmission timestamp and the reception timestamp of two or more synthetic packets; and deriving an available bandwidth of the network segment based on the calculated inter-packet delay variation.
- the derived available bandwidth is used to trigger one or more actions, such as the action of changing the configuration of traffic shapers for the network segment.
- the deriving of the available bandwidth preferably takes into account one or more previously calculated inter packet delay variations.
- the plurality of synthetic packets may have the same predetermined number of bytes, and may be transmitted substantially consecutively.
- One implementation for calculating the inter packet delay variation comprises calculating a first delay as a difference between the reception timestamp and the transmission timestamp for a first of the plurality of synthetic packets; calculating a second delay for a second synthetic packet as a difference between the reception timestamp and the transmission timestamp for the second of the plurality of synthetic packets, and setting the inter packet delay variation as the difference between the second delay and the first delay.
- the available bandwidth of the network segment may be determined by multiplying a number of bytes in a first synthetic packet by the rate of a first link in the network segment and adding the calculated inter packet delay variation for the first synthetic packet, and dividing the resulting sum by a number of bytes in the synthetic packet.
- FIG. 1 illustrates a generic Ethernet network
- FIG. 2 illustrates an Ethernet network with a radio link.
- FIG. 2 a illustrates measurement packets over a radio link of an Ethernet network.
- FIG. 3 illustrates how the CIR values for a plurality of classes of services evolve in reaction to changes in the total available bandwidth of a link.
- FIG. 4 illustrates a generic Ethernet network in a ring topology.
- FIG. 5 illustrates an Ethernet network in a ring topology with bandwidth adjustments resulting from measurements taken at each node in the Ethernet ring.
- FIG. 6 illustrates an Ethernet network with a pair of rings with bandwidth adjustments resulting from measurements taken at each node in each of the pair of rings.
- the embodiment described herein allows measuring the available bandwidth in a network segment 104 between a first 101 and second 102 location in a network 100 .
- the traffic is generated by a data source device 110 and received at a data sink device 120 .
- a Traffic Shaper and Scheduler (TSS) 103 function is implemented to shape the user traffic to be compliant to the configured CIR assigned to the connections from the source device 110 .
- a Synthetic Traffic Generator (STG) 105 is implemented downstream and in the path of the traffic from the TSS 103 on the same side as the data source device 110 relative to the position of the network 100 .
- the STG is implemented at the first location 101 .
- the TSS 103 and STG 105 can be combined at the first location 101 . They could be, for example integrated as part of a Network Interface Device (NID) or an access switch. As another option, the TSS 103 and STG 105 can be integrated as part of the Data Source Device 110 and together forming the first location 101 .
- NID Network Interface Device
- the TSS 103 and STG 105 can be integrated as part of the Data Source Device 110 and together forming the first location 101 .
- a bandwidth measurement function (BMF) 109 is implemented at a second location 102 , on the opposite side relative to the position of the network 100 and upstream from the data sink device 120 .
- BMF bandwidth measurement function
- two or more synthetic packets (not user traffic) are generated by a STG 105 and transmitted at the highest possible rate and packet priority and without any substantial gap to the BMF 109 across the network 100 with the goal of obtaining measurement for the purpose of measuring the available bandwidth on the network 100 between the first location 101 and the second 102 location.
- the synthetic packets are discarded.
- the BMF 109 is installed in parallel to the sink device 120 and is not in the direct data path between the network 100 and the sink device 120 .
- the BMF can also be implemented a part of the sink device 120 or within a NID or other network elements.
- IPDV Inter Packet Delay Variation
- BDF Bandwidth Decision Function
- the packet is transmitted into the segment using a first link 106 .
- the packet exits the segment using a second link 107 .
- the BDF can be collocated with the BMF 109 at the second location, or the BDF 115 can be implemented at a separate location and used in a centralized manner, so multiple bandwidth measurements from different points in the network can be analyzed together.
- the BDF 115 can trigger one or more actions based on the available bandwidth measured. For example, if a degradation or improvement of available bandwidth is estimated, an adjustment to the CIR and EIR settings for each class of service may be required. If an adjustment is needed to the settings (CIR and/or EIR) for one or more class of service for the network 100 , the BDF 115 communicates the new settings to the TSS 103 . In another embodiment, traffic belonging to one or more Class of Service (CoS) may be moved to a different network path with better performance.
- CoS Class of Service
- FIG. 2 illustrates the case where the network segment is the radio link 202 created by a pair of radio transceivers ( 220 , 230 ), in this case the first and second links are the same.
- measuring of the available bandwidth achieved by the pair of radio transceivers ( 220 , 230 ) is important to detect any variance in the available bandwidth as the performance of the radio link may be negatively impacted by a number of external factors such as: a source of interference, weather conditions or modulation adjustments between the radio transceivers ( 220 , 230 ).
- the first and second locations are implemented close to the radio transceivers.
- the synthetic traffic from the STG 105 flows over link 201 to a first radio transceiver 220 where it may experience internal congestion before the synthetic packet is transmitted over link 202 toward a second radio transceiver 230 where the synthetic packet may further experience internal congestion before it is transmitted over link 203 to the BMF 109 .
- the roles of STG 105 and of BMF 109 may be combined into the same device, and a pair of these combined devices would replace devices 105 and 109 in order to allow for the measuring of the bandwidth over link 202 in both directions (from link 201 to link 203 and from link 203 to link 201 ).
- the synthetic packets generated by the STG 105 may be constructed to prevent data compression in order to achieve more accurate IPDV measurements over the radio Link 202 . If the synthetic packets are compressed over the radio Link 202 , the content of each of the synthetic packets should ideally result in the same level of compression to limit any significant discrepancy in the IPDV measurements.
- the IPDV is calculated by the BMF 109 and used by the BDF 115 to derive the available bandwidth of a link in a specific direction since links (especially radios, but also xDSL and others) may deliver asymmetric bandwidth for each direction of the link.
- FIG. 2A shows the variables used in an example IPDV measurement using two synthetic packets in the radio link example. Packet A and Packet B of Lengths L A and L B , respectively.
- synthetic packet A is generated by the STF 105 at time T 0 and received at the BMF 109 at time RX A .
- the packet is marked with transmission timestamp T 0 and reception timestamp RX a .
- the STG 105 Immediately after transmitting synthetic packet A, the STG 105 generates synthetic packet B at time T 1 , and synthetic packet B is received at the BMF 109 at time RX B .
- D B S 1B +C 1B +S WB +S 2B +OFF
- IPDV L B *R W ⁇ L B *R 1
- the BDF 115 may use hysteresis or weighted average techniques to smooth the results to avoid reacting on transient variations. This is particularly useful for links that may experience periods of instability (such as radio links). IPDV measurements may also be considered as statistical outliers, and ignored or deleted, if they are significantly outside of the range of previous measurements.
- the IPDV calculation (or average) can be used to estimate the bandwidth available between the first and second locations at any given moment using:
- R W ( IPDV+L B R 1 )/ L B
- the bandwidth available R W represents the current bandwidth on the radio link.
- R W is the rate of the first link 106 .
- the total available bandwidth 300 of a link 100 is divided amongst a plurality of classes of service with the corresponding CIR configured by the traffic shaper and scheduler 103 .
- CIR 1 301 is associated with Class of Service 1
- CIR 2 302 is associated with Class of Service 2
- CIR 3 303 is associated with Class of Service 3
- the remaining bandwidth is associated with a best effort class of service 304 .
- the shaping can be done for an aggregate of all connections sharing a class of service or implemented for each connection separately.
- the link 100 may experience degradation in the total available bandwidth 300 it can deliver in the measured direction.
- the BDF 115 takes action to reduce the CIR 3 303 to zero (effectively dropping all user traffic for that class of service). It can also take action to eliminate some or all of the best effort traffic 304 .
- the BDF 115 takes action to communicate with the TSS 103 and restore the value for CIR 3 303 to its originally defined value and the amount of remaining bandwidth 304 for best effort traffic is also restored to its originally defined value.
- the BDF 115 reduces the CIR values ( 301 , 302 , 303 , 304 ) for each class of service prorated to its original percentage of the total allocated bandwidth 300 and will reinstate the original CIR values ( 301 , 302 , 303 , 304 ) for each class of service using the same prorated technique when more bandwidth becomes available on link 100 .
- the BDF 115 needs to keep track of the number of classes supported, applicable in the direction of the IPDV obtained for a link with the goal to avoid impacting the CIR of the higher priority classes of traffic when the overall performance of the link R W (in the measured direction) is degraded compared to the previous IPDV measurement.
- the estimate of the R W measurement can be smoothed using different known techniques such as weighted averaging, moving averages etc.
- the smoothed estimate SR W can then be compared to different thresholds for the BDF to decide which action to take.
- the available bandwidth determined from the IPDV can be allocated in such a way that the CIR for the link corresponds to the CIR required to support the class with the highest priority of service, and with the second class configured to use the remaining bandwidth in a best-effort mode.
- the CIR for the highest class of traffic is protected as long as the achievable bandwidth on the link 100 is sufficient to meet that CIR.
- the CIR values for each class of service are adjusted (if needed), generally beginning with the CIR associated with the least priority class in order to minimize any impact to the higher classes of service.
- the BDF 115 can also take action to report any changes to the IPDV and resulting bandwidth and CIR settings to the network management entity, to allow full reporting of the changes.
- the BDF 115 may use well-known hysteresis techniques to limit the number of notifications while retaining the ability to report on drastic and rapid changes when a link is highly unstable.
- Networks are now generally designed with redundant links to recover from link failures.
- a network consisting of individual links ( 401 , 402 , 403 and 404 ) installed between Ring nodes ( 410 , 411 , 412 and 413 ) is deployed in a ring topology to offer redundant data path in case one of the link ( 401 , 402 , 403 and 404 ) fails.
- the physical ring topology needs to be logically broken by disabling one of the links ( 401 , 402 , 403 and 404 ) where the disabled link ( 404 in FIG. 4 ) is known as the Ring Protection Link (RPL) 420 .
- RPL Ring Protection Link
- Traffic from a Data Source Device 110 flows toward a Data Sink Device 120 by transiting via ring node 413 toward ring node 412 over link 403 and then from ring node 412 to ring node 410 over link 402 before being extracted from the ring by ring node 410 and being passed to the Data Sink Device 120 .
- Traffic in the opposite direction, from the Data Sink Device 120 to Data Source Device 110 transits from ring node 410 to ring node 412 over Link 402 and then from ring node 412 to ring node 413 over Link 403 and finally from ring node 413 to Data Source Device 110 .
- each ring node incorporates a TSS ( 103 a, 103 b, 103 c and 103 d ), a STG ( 105 a, 105 b, 105 c and 105 d ) and a BMF ( 109 a, 109 b , 109 c and 109 d ).
- ring node 510 also incorporates a BDF 115 that centrally collects all of the IPDV measurements for each direction of each Ethernet Link ( 401 , 402 , 403 and 404 ) to maintain a global understanding of the total achievable bandwidth over the network 100 .
- the BDF 115 in ring node 510 Rather than waiting for the total failure of a Link to reconfigure a Ring topology, it is desirable for the BDF 115 in ring node 510 to react to the degradation of the performance of a Link ( 401 , 402 , 403 or 404 ) by leveraging the IPDV measurements for each Link ( 401 , 402 , 403 and 404 ) and in each direction and adjusting the value of the CIR and EIR for each class of services for each Link ( 401 , 402 , 403 and 404 ) of the network 100 and enforced by the TSS ( 103 a, 103 b , 103 c, 103 d ) of each ring node ( 510 , 511 , 512 , 513 ).
- the BDF 115 in ring node 510 effects one or more actions such as CIR adjustments to one or more links ( 401 , 402 , 403 and 404 ) to be implemented by the TSS ( 103 a, 103 b, 103 c, 103 d ) by moving the RPL 520 to the degraded link, disabling the degraded link and enabling the use of the now available and fully functional link ( 404 in FIG. 5 ) to achieve the best possible overall performance on the network 100 .
- the severe degradation or the total failure of link 402 results in the BDF 115 moving the location of the RPL 520 from link 404 to link 402 .
- the BDF 115 may use the IPDV measurements (or the total available bandwidth 300 ) to determine where the RPL should be located, possibly reverting the RPL 520 to its original location over link 404 .
- the Bandwidth Decision Node 115 may be distributed among a plurality of ring nodes ( 510 , 511 , 512 and 513 ) rather than being integrated in a single centralized ring node ( 510 as per FIG. 5 ).
- FIG. 6 illustrates a network 100 configured with a plurality of rings.
- the first ring consists of virtual Local Area Networks (VLANs) 611 , 612 , 613 and 614 with the RPL 610 assigned to Link 604 (VLAN 614 ).
- the second ring consists of virtual LANs 621 , 622 , 623 and 624 with the RPL 620 assigned to link 603 (VLAN 623 ).
- a dual-ring topology can serve various purposes, including supporting multiple Data Source Devices ( 640 and 642 ) and multiple Data Sink Devices ( 641 and 643 ) where each pair of devices ( 640 and 641 , 642 and 643 ) is connected over a different ring.
- Ring node 510 incorporates a centralized BDF 115 that is responsible for collecting and analysis of the IPDV measurements collected by each of the Ring nodes ( 510 , 511 , 512 and 513 ) in each direction of each link ( 601 , 602 , 603 and 604 ).
- the BDF 115 is also responsible to take action to trigger adjustments to the CIR and EIR values to be implemented by the TSS ( 103 a, 103 b, 103 c , 103 d ) for each VLAN and for each class of service defined for VLANs 611 , 612 , 613 and 614 for the first ring and VLANs 621 , 622 , 623 and 624 for the second ring as well as the determination of where the RPL ( 610 and 620 ) should be set for each ring.
- the degradation of a link present on more than one ring for example 601 and 602 in FIG.
- the TSS may require adjustments to the CIR and EIR values for each class of service enforced by the TSS ( 103 a, 103 b, 103 c , 103 d ) on each impacted ring and for other links feeding into or fed from a degraded link ( 601 , 602 , 603 , 604 ).
- the BDF 115 may take action to disable the degraded link and, optionally, locate both RPLs ( 610 and 620 ) on the same link ( 601 , 602 , 603 , and 604 ) in order to achieve the best possible overall bandwidth.
Abstract
A method and system are provided for measuring the bandwidth of a network segment between a first location and a second location by generating a plurality of synthetic packets at the first location and marking each generated synthetic packet with a transmission timestamp indicative of a first clock time at which the synthetic packet is generated; transmitting each of the generated synthetic packets to the second location; receiving each synthetic packet at the second location and marking each received synthetic packet with a reception timestamp indicative of a second clock time at which the synthetic packet is received at the second location; calculating an inter-packet delay variation of the network segment based on the transmission timestamp and the reception timestamp of two or more synthetic packets; and deriving an available bandwidth of the network segment based on the calculated inter-packet delay variation.
Description
- This invention is directed towards the ability to react to a detected change in the available bandwidth of a sub-rate link, for instance wireless, by adjusting operating parameters for the link with regard to service assurance, traffic engineering, service reliability and related functions.
- Parameters and settings for service assurance, traffic engineering and service reliability (amongst a plurality of other network services), are often determined and defined based on the bandwidth of a link. This is the case for committed information rate (CIR) and excess information rate (EIR) settings of an Ethernet virtual circuit (EVC). Time synchronization protocols such as IEEE 1588v2 also require a link with a stable bandwidth in order to derive and maintain clock synchronization between a pair of networking devices. Flow control mechanisms have been used to adapt to the bandwidth of a link or network segment (ref: U.S. Pat. No. 8,139,485). Such flow control protocols require the knowledge of the protocol at each node in the segment.
- By being able to detect and measure the available bandwidth of a link or network segment, it is possible to react to a change in the available bandwidth and adjust impacted parameters (for instance CIR and EIR) or adjust the clock synchronization between a pair of networking devices in order to attempt preserving the quality of the services offered over a link when its bandwidth is negatively impacted by spurious or permanent degradation, spurious degradation being sometimes common for wireless links suffering from temporary weather impairments and external sources of interference.
- In accordance with one embodiment, a method and system are provided for measuring the bandwidth of a network segment between a first location and a second location by generating a plurality of synthetic packets at the first location and marking each generated synthetic packet with a transmission timestamp indicative of a first clock time at which the synthetic packet is generated; transmitting each of the generated synthetic packets to the second location; receiving each synthetic packet at the second location and marking each received synthetic packet with a reception timestamp indicative of a second clock time at which the synthetic packet is received at the second location; calculating an inter-packet delay variation of the network segment based on the transmission timestamp and the reception timestamp of two or more synthetic packets; and deriving an available bandwidth of the network segment based on the calculated inter-packet delay variation.
- In one implementation, the derived available bandwidth is used to trigger one or more actions, such as the action of changing the configuration of traffic shapers for the network segment. The deriving of the available bandwidth preferably takes into account one or more previously calculated inter packet delay variations. The plurality of synthetic packets may have the same predetermined number of bytes, and may be transmitted substantially consecutively.
- One implementation for calculating the inter packet delay variation comprises calculating a first delay as a difference between the reception timestamp and the transmission timestamp for a first of the plurality of synthetic packets; calculating a second delay for a second synthetic packet as a difference between the reception timestamp and the transmission timestamp for the second of the plurality of synthetic packets, and setting the inter packet delay variation as the difference between the second delay and the first delay. The available bandwidth of the network segment may be determined by multiplying a number of bytes in a first synthetic packet by the rate of a first link in the network segment and adding the calculated inter packet delay variation for the first synthetic packet, and dividing the resulting sum by a number of bytes in the synthetic packet.
- The invention may best be understood by reference to the following description taken in conjunction with the accompanying drawings.
-
FIG. 1 illustrates a generic Ethernet network. -
FIG. 2 illustrates an Ethernet network with a radio link. -
FIG. 2 a illustrates measurement packets over a radio link of an Ethernet network. -
FIG. 3 illustrates how the CIR values for a plurality of classes of services evolve in reaction to changes in the total available bandwidth of a link. -
FIG. 4 illustrates a generic Ethernet network in a ring topology. -
FIG. 5 illustrates an Ethernet network in a ring topology with bandwidth adjustments resulting from measurements taken at each node in the Ethernet ring. -
FIG. 6 illustrates an Ethernet network with a pair of rings with bandwidth adjustments resulting from measurements taken at each node in each of the pair of rings. - Although the invention will be described in connection with certain preferred embodiments, it will be understood that the invention is not limited to those particular embodiments. On the contrary, the invention is intended to cover all alternatives, modifications, and equivalent arrangements as may be included within the spirit and scope of the invention as defined by the appended claims.
- Turning now to the drawings and referring first to
FIG. 1 , the embodiment described herein allows measuring the available bandwidth in a network segment 104 between a first 101 and second 102 location in anetwork 100. The traffic is generated by adata source device 110 and received at adata sink device 120. Betweendevices source device 110. Furthermore, a Synthetic Traffic Generator (STG) 105 is implemented downstream and in the path of the traffic from theTSS 103 on the same side as thedata source device 110 relative to the position of thenetwork 100. The STG is implemented at thefirst location 101. Optionally, the TSS 103 and STG 105 can be combined at thefirst location 101. They could be, for example integrated as part of a Network Interface Device (NID) or an access switch. As another option, the TSS 103 and STG 105 can be integrated as part of theData Source Device 110 and together forming thefirst location 101. - A bandwidth measurement function (BMF) 109 is implemented at a
second location 102, on the opposite side relative to the position of thenetwork 100 and upstream from thedata sink device 120. At regular intervals, two or more synthetic packets (not user traffic) are generated by aSTG 105 and transmitted at the highest possible rate and packet priority and without any substantial gap to theBMF 109 across thenetwork 100 with the goal of obtaining measurement for the purpose of measuring the available bandwidth on thenetwork 100 between thefirst location 101 and the second 102 location. After analysis by the BMF 109, the synthetic packets are discarded. In another embodiment, the BMF 109 is installed in parallel to thesink device 120 and is not in the direct data path between thenetwork 100 and thesink device 120. The BMF can also be implemented a part of thesink device 120 or within a NID or other network elements. - An Inter Packet Delay Variation (IPDV) is calculated by the
BMF 109 from the synthetic packets and is made available to a Bandwidth Decision Function (BDF) 115 that uses the IPDV to determine the available bandwidth of the network segment 104 between the first and second location. The packet is transmitted into the segment using a first link 106. The packet exits the segment using a second link 107. The BDF can be collocated with theBMF 109 at the second location, or the BDF 115 can be implemented at a separate location and used in a centralized manner, so multiple bandwidth measurements from different points in the network can be analyzed together. - The BDF 115 can trigger one or more actions based on the available bandwidth measured. For example, if a degradation or improvement of available bandwidth is estimated, an adjustment to the CIR and EIR settings for each class of service may be required. If an adjustment is needed to the settings (CIR and/or EIR) for one or more class of service for the
network 100, the BDF 115 communicates the new settings to theTSS 103. In another embodiment, traffic belonging to one or more Class of Service (CoS) may be moved to a different network path with better performance. -
FIG. 2 illustrates the case where the network segment is theradio link 202 created by a pair of radio transceivers (220, 230), in this case the first and second links are the same. In this case, measuring of the available bandwidth achieved by the pair of radio transceivers (220, 230) is important to detect any variance in the available bandwidth as the performance of the radio link may be negatively impacted by a number of external factors such as: a source of interference, weather conditions or modulation adjustments between the radio transceivers (220, 230). To obtain an estimate of the radio bandwidth, the first and second locations are implemented close to the radio transceivers. For the purpose of obtaining bandwidth measurements, the synthetic traffic from theSTG 105 flows overlink 201 to afirst radio transceiver 220 where it may experience internal congestion before the synthetic packet is transmitted overlink 202 toward asecond radio transceiver 230 where the synthetic packet may further experience internal congestion before it is transmitted overlink 203 to theBMF 109. - In another embodiment, the roles of
STG 105 and of BMF 109 may be combined into the same device, and a pair of these combined devices would replacedevices link 202 in both directions (fromlink 201 tolink 203 and fromlink 203 to link 201). - When the pair of radio transceivers (220, 230) supports data compression, the synthetic packets generated by the
STG 105 may be constructed to prevent data compression in order to achieve more accurate IPDV measurements over theradio Link 202. If the synthetic packets are compressed over theradio Link 202, the content of each of the synthetic packets should ideally result in the same level of compression to limit any significant discrepancy in the IPDV measurements. - The IPDV is calculated by the
BMF 109 and used by theBDF 115 to derive the available bandwidth of a link in a specific direction since links (especially radios, but also xDSL and others) may deliver asymmetric bandwidth for each direction of the link. -
FIG. 2A shows the variables used in an example IPDV measurement using two synthetic packets in the radio link example. Packet A and Packet B of Lengths LA and LB, respectively. - Looking at
FIG. 2 in more detail, synthetic packet A is generated by theSTF 105 at time T0 and received at theBMF 109 at time RXA. The packet is marked with transmission timestamp T0 and reception timestamp RXa. - It is known that:
-
RX A =T 0 +D A -
D A =S 1A +C 1A +S WA +S 2A+OFF -
- Where
- S1A=
Serialisation delay 130 on link 1 (201) - C1A=
Congestion delay 131 inradio 220 - R1=Rate on link 1 (201)
- RW=Rate on
wireless link 202 - SWA=
Serialization delay 132 onwireless link 202 - S2A=
Serialization delay 133 on link 2 (203) - OFF=Clock offset at
bandwidth measuring node 109 when compared to the clock of theSTG 105
- Immediately after transmitting synthetic packet A, the
STG 105 generates synthetic packet B at time T1, and synthetic packet B is received at theBMF 109 at time RXB. - It is known that:
-
RX B =T 1 +D B -
D B =S 1B +C 1B +S WB +S 2B+OFF -
- Where
- S1B=
Serialisation delay 134 onlink 1 201 - C1B=
Congestion delay 135 inradio 220=C1A+(LA*RW)−S1B - RW=Rate on
wireless link 202 - SWB=
Serialization delay 136 onwireless link 202 - S2B=
Serialization delay 137 on link 2 (203) - OFF=Clock offset at
bandwidth measuring node 109 when compared to the clock of theSTG 105 - IPDV is computed as:
-
IPDV=(RX B −T 1)−(RX A −T 0)=D B −D A -
- Assuming LB=LA
-
IPDV=L B *R W −L B *R 1 - When using the IPDV measurements, the
BDF 115 may use hysteresis or weighted average techniques to smooth the results to avoid reacting on transient variations. This is particularly useful for links that may experience periods of instability (such as radio links). IPDV measurements may also be considered as statistical outliers, and ignored or deleted, if they are significantly outside of the range of previous measurements. - The IPDV calculation (or average) can be used to estimate the bandwidth available between the first and second locations at any given moment using:
-
R W=(IPDV+L B R 1)/L B - If the first and second locations are located near two radio transceivers, the bandwidth available RW represents the current bandwidth on the radio link.
- The above calculation applies similarly to measure the available bandwidth of a network segment. In this case RW is the rate of the first link 106.
- Looking now at
FIG. 3 , the totalavailable bandwidth 300 of alink 100 is divided amongst a plurality of classes of service with the corresponding CIR configured by the traffic shaper andscheduler 103.CIR 1 301 is associated with Class ofService 1,CIR 2 302 is associated with Class ofService 2,CIR 3 303 is associated with Class ofService 3 and the remaining bandwidth is associated with a best effort class ofservice 304. The shaping can be done for an aggregate of all connections sharing a class of service or implemented for each connection separately. - Over time, the
link 100 may experience degradation in the totalavailable bandwidth 300 it can deliver in the measured direction. In this case theBDF 115 takes action to reduce theCIR 3 303 to zero (effectively dropping all user traffic for that class of service). It can also take action to eliminate some or all of thebest effort traffic 304. After another period of time, when the measurements indicate that the bandwidth available between the first and second locations is increased to a given level, theBDF 115 takes action to communicate with theTSS 103 and restore the value forCIR 3 303 to its originally defined value and the amount of remainingbandwidth 304 for best effort traffic is also restored to its originally defined value. In another embodiment, theBDF 115 reduces the CIR values (301, 302, 303, 304) for each class of service prorated to its original percentage of the total allocatedbandwidth 300 and will reinstate the original CIR values (301, 302, 303, 304) for each class of service using the same prorated technique when more bandwidth becomes available onlink 100. - The
BDF 115 needs to keep track of the number of classes supported, applicable in the direction of the IPDV obtained for a link with the goal to avoid impacting the CIR of the higher priority classes of traffic when the overall performance of the link RW (in the measured direction) is degraded compared to the previous IPDV measurement. - The estimate of the RW measurement can be smoothed using different known techniques such as weighted averaging, moving averages etc. The smoothed estimate SRW can then be compared to different thresholds for the BDF to decide which action to take.
- For example, if there are two classes of traffic defined over the
link 100 and enforced by theTSS 103, the available bandwidth determined from the IPDV can be allocated in such a way that the CIR for the link corresponds to the CIR required to support the class with the highest priority of service, and with the second class configured to use the remaining bandwidth in a best-effort mode. This way, the CIR for the highest class of traffic is protected as long as the achievable bandwidth on thelink 100 is sufficient to meet that CIR. When there are more than two classes of service, the CIR values for each class of service are adjusted (if needed), generally beginning with the CIR associated with the least priority class in order to minimize any impact to the higher classes of service. - Whenever the CIR for a class of service is modified, adequate notifications can be sent to a network management entity or a similar device to alert the network manager or the operator of the condition.
- Furthermore, the
BDF 115 can also take action to report any changes to the IPDV and resulting bandwidth and CIR settings to the network management entity, to allow full reporting of the changes. In order to avoid the generation of a large number of notifications, theBDF 115 may use well-known hysteresis techniques to limit the number of notifications while retaining the ability to report on drastic and rapid changes when a link is highly unstable. - Networks are now generally designed with redundant links to recover from link failures. Looking at
FIG. 4 , a network consisting of individual links (401, 402, 403 and 404) installed between Ring nodes (410, 411, 412 and 413) is deployed in a ring topology to offer redundant data path in case one of the link (401, 402, 403 and 404) fails. As is well known in the art, the physical ring topology needs to be logically broken by disabling one of the links (401, 402, 403 and 404) where the disabled link (404 inFIG. 4 ) is known as the Ring Protection Link (RPL) 420. Traffic from aData Source Device 110 flows toward aData Sink Device 120 by transiting viaring node 413 towardring node 412 overlink 403 and then fromring node 412 toring node 410 overlink 402 before being extracted from the ring byring node 410 and being passed to theData Sink Device 120. Traffic in the opposite direction, from theData Sink Device 120 toData Source Device 110, transits fromring node 410 toring node 412 overLink 402 and then fromring node 412 toring node 413 overLink 403 and finally fromring node 413 toData Source Device 110. In the advent of a failure ofEthernet Link RPL 420 will be moved to the failed Link (402 or 403), and any traffic betweenData Source Device 110 andData Sink Device 120 would flow overLinks ring nodes - Looking now at
FIG. 5 , each ring node (510, 511, 512 and 513) incorporates a TSS (103 a, 103 b, 103 c and 103 d), a STG (105 a, 105 b, 105 c and 105 d) and a BMF (109 a, 109 b, 109 c and 109 d). Furthermore,ring node 510 also incorporates aBDF 115 that centrally collects all of the IPDV measurements for each direction of each Ethernet Link (401, 402, 403 and 404) to maintain a global understanding of the total achievable bandwidth over thenetwork 100. Rather than waiting for the total failure of a Link to reconfigure a Ring topology, it is desirable for theBDF 115 inring node 510 to react to the degradation of the performance of a Link (401, 402, 403 or 404) by leveraging the IPDV measurements for each Link (401, 402, 403 and 404) and in each direction and adjusting the value of the CIR and EIR for each class of services for each Link (401, 402, 403 and 404) of thenetwork 100 and enforced by the TSS (103 a, 103 b, 103 c, 103 d) of each ring node (510, 511, 512, 513). If the IPDV measurements indicate that the degradation of a Link (401, 402, 403 or 404) is beyond a predetermined threshold, theBDF 115 inring node 510 effects one or more actions such as CIR adjustments to one or more links (401, 402, 403 and 404) to be implemented by the TSS (103 a, 103 b, 103 c, 103 d) by moving theRPL 520 to the degraded link, disabling the degraded link and enabling the use of the now available and fully functional link (404 inFIG. 5 ) to achieve the best possible overall performance on thenetwork 100. For instance, the severe degradation or the total failure oflink 402 results in theBDF 115 moving the location of theRPL 520 fromlink 404 to link 402. Upon restoration of theavailable bandwidth 300 of a link (401, 402, 403 and 404) to its original value (or to another value that is high enough), theBDF 115 may use the IPDV measurements (or the total available bandwidth 300) to determine where the RPL should be located, possibly reverting theRPL 520 to its original location overlink 404. - In an alternate embodiment, the
Bandwidth Decision Node 115 may be distributed among a plurality of ring nodes (510, 511, 512 and 513) rather than being integrated in a single centralized ring node (510 as perFIG. 5 ). -
FIG. 6 illustrates anetwork 100 configured with a plurality of rings. The first ring consists of virtual Local Area Networks (VLANs) 611, 612, 613 and 614 with theRPL 610 assigned to Link 604 (VLAN 614). The second ring consists ofvirtual LANs RPL 620 assigned to link 603 (VLAN 623). A dual-ring topology can serve various purposes, including supporting multiple Data Source Devices (640 and 642) and multiple Data Sink Devices (641 and 643) where each pair of devices (640 and 641, 642 and 643) is connected over a different ring. Using multiple rings may also be useful to support multiple BGP routers in a backhaul network where the traffic to the Internet is load balanced between the plurality of BGP routers to achieve higher bandwidth and to offer redundancy in case one of the BGP routers fails. Server redundancy and load balancing may also be achieved with a plurality of rings.Ring node 510 incorporates acentralized BDF 115 that is responsible for collecting and analysis of the IPDV measurements collected by each of the Ring nodes (510, 511, 512 and 513) in each direction of each link (601, 602, 603 and 604). TheBDF 115 is also responsible to take action to trigger adjustments to the CIR and EIR values to be implemented by the TSS (103 a, 103 b, 103 c, 103 d) for each VLAN and for each class of service defined forVLANs VLANs FIG. 6 ), may require adjustments to the CIR and EIR values for each class of service enforced by the TSS (103 a, 103 b, 103 c, 103 d) on each impacted ring and for other links feeding into or fed from a degraded link (601, 602, 603, 604). - When the IPDV measurements indicate the degradation of a single link (601, 602, 603, 604), the
BDF 115 may take action to disable the degraded link and, optionally, locate both RPLs (610 and 620) on the same link (601, 602, 603, and 604) in order to achieve the best possible overall bandwidth. - While particular embodiments and applications of the present invention have been illustrated and described, it is to be understood that the invention is not limited to the precise construction and compositions disclosed herein and that various modifications, changes, and variations may be apparent from the foregoing descriptions without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (16)
1. A method for measuring the bandwidth of a network segment between a first location and a second location, comprising
generating a plurality of synthetic packets at the first location and marking each generated synthetic packet with a transmission timestamp indicative of a first clock time at which the synthetic packet is generated,
transmitting each of the generated synthetic packets said second location,
receiving each synthetic packet at the second location and marking each received synthetic packet with a reception timestamp indicative of a second clock time at which the synthetic packet is received at the second location,
calculating an inter-packet delay variation of the network segment based on the transmission timestamp and the reception timestamp of two or more synthetic packets, and
deriving an available bandwidth of the network segment based on the calculated inter-packet delay variation.
2. The method of claim 1 wherein said derived available bandwidth is used to trigger one or more actions.
3. The method of claim 2 wherein said one or more actions comprises changing the configuration of traffic shapers for said network segment.
4. The method of claim 1 wherein the deriving of said available bandwidth takes into account one or more previously calculated inter packet delay variations.
5. The method of claim 1 wherein each of said plurality of synthetic packets has the same predetermined number of bytes.
6. The method of claim 1 wherein said plurality of synthetic packets are transmitted substantially consecutively.
7. The method of claim 1 , in which said calculating the inter packet delay variation comprises
calculating a first delay as a difference between the reception timestamp and the transmission timestamp for a first of said plurality of synthetic packets,
calculating a second delay for a second synthetic packet as a difference between the reception timestamp and the transmission timestamp for the second of said plurality of synthetic packets, and
setting the inter packet delay variation as the difference between said second delay and said first delay.
8. A method according to claim 1 , in which said deriving an available bandwidth of the network segment comprises
multiplying a number of bytes in a first synthetic packet by the rate of a first link in said network segment and adding the calculated inter packet delay variation for said first synthetic packet, and
dividing the resulting sum by a number of bytes in said synthetic packet.
9. A system for measuring the bandwidth of a network segment between a first location and a second location, said system comprising
a first processor adapted to
generate a plurality of synthetic packets at the first location and mark each generated synthetic packet with a transmission timestamp indicative of a first clock time at which the synthetic packet is generated, and
transmit each of the generated synthetic to said second location, and
a second processor adapted to
receive each synthetic packet at the second location and mark each received synthetic packet with a reception timestamp indicative of a second clock time at which the synthetic packet is received at the second location,
calculate an inter-packet delay variation of the network segment based on the transmission timestamp and the reception timestamp of two or more synthetic packets, and
derive an available bandwidth of the network segment based on the calculated inter-packet delay variation.
10. The system of claim 9 in which said second processor is adapted to use said derived available bandwidthto trigger one or more actions.
11. The system of claim 10 wherein said one or more actions comprises changing the configuration of traffic shapers for said network segment.
12. The system of claim 9 in which said second processor is adapted to derive said available bandwidth based at least in part on one or more previously calculated inter packet delay variations.
13. The system of claim 9 in which said first processor is adapted to generate each of said plurality of synthetic packets with the same predetermined number of bytes.
14. The system of claim 9 in which said first processor is adapted to transmit said plurality of synthetic packets substantially consecutively.
15. The system of claim 9 , in which said second processor is adapted to calculate the inter packet delay variation by
calculating a first delay as a difference between the reception timestamp and the transmission timestamp for a first of said plurality of synthetic packets,
calculating a second delay for a second synthetic packet as a difference between the reception timestamp and the transmission timestamp for the second of said plurality of synthetic packets, and
setting the inter packet delay variation as the difference between said second delay and said first delay.
16. A system according to claim 9 , in which said second processor is adapted to derive an available bandwidth of the network segment by
multiplying a number of bytes in a first synthetic packet by the rate of a first link in said network segment and adding the calculated inter packet delay variation for said first synthetic packet, and
dividing the resulting sum by a number of bytes in said synthetic packet.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/706,420 US20140160945A1 (en) | 2012-12-06 | 2012-12-06 | Using bandwidth measurements to adjust cir and eir on a sub-rate link |
US15/643,027 US10484291B2 (en) | 2012-12-06 | 2017-07-06 | Using bandwidth measurements to adjust CIR and EIR on a sub-rate link |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/706,420 US20140160945A1 (en) | 2012-12-06 | 2012-12-06 | Using bandwidth measurements to adjust cir and eir on a sub-rate link |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/643,027 Continuation US10484291B2 (en) | 2012-12-06 | 2017-07-06 | Using bandwidth measurements to adjust CIR and EIR on a sub-rate link |
Publications (1)
Publication Number | Publication Date |
---|---|
US20140160945A1 true US20140160945A1 (en) | 2014-06-12 |
Family
ID=50880877
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US13/706,420 Abandoned US20140160945A1 (en) | 2012-12-06 | 2012-12-06 | Using bandwidth measurements to adjust cir and eir on a sub-rate link |
US15/643,027 Active 2033-02-07 US10484291B2 (en) | 2012-12-06 | 2017-07-06 | Using bandwidth measurements to adjust CIR and EIR on a sub-rate link |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/643,027 Active 2033-02-07 US10484291B2 (en) | 2012-12-06 | 2017-07-06 | Using bandwidth measurements to adjust CIR and EIR on a sub-rate link |
Country Status (1)
Country | Link |
---|---|
US (2) | US20140160945A1 (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150236962A1 (en) * | 2014-02-14 | 2015-08-20 | Exinda Networks PTY, Ltd. of Australia | Method and system for using dynamic bandwidth detection to drive quality of service control refinement |
WO2016149866A1 (en) * | 2015-03-20 | 2016-09-29 | 华为技术有限公司 | Method and device for measuring available bandwidth in end to end path |
US20180150432A1 (en) * | 2015-08-31 | 2018-05-31 | Huawei Technologies Co., Ltd. | Method and apparatus for measuring time samp unit of remote device |
US20190230001A1 (en) * | 2014-04-17 | 2019-07-25 | Accedian Networks Inc. | System and method for out-of-line real-time in-service performance measurement |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9735874B2 (en) * | 2012-07-18 | 2017-08-15 | Accedian Networks Inc. | Programmable small form-factor pluggable module |
CN107920028B (en) * | 2017-11-14 | 2020-06-16 | 东软集团股份有限公司 | Data packet sending method, device and equipment |
US10630539B2 (en) * | 2018-08-07 | 2020-04-21 | International Business Machines Corporation | Centralized rate limiters for services in cloud based computing environments |
US11190983B2 (en) * | 2019-04-22 | 2021-11-30 | T-Mobile Usa, Inc. | Network latency control |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001091375A2 (en) * | 2000-05-19 | 2001-11-29 | Williams Communications Llc | Method and system for measuring one-way delay variation |
US20060215574A1 (en) * | 2005-03-24 | 2006-09-28 | Microsoft Corporation | Bandwidth estimation in broadband access networks |
US20070081561A1 (en) * | 2005-10-11 | 2007-04-12 | International Business Machines Corporation | Single ended solution for estimation of bandwidth delay product |
US20070217448A1 (en) * | 2006-03-16 | 2007-09-20 | Microsoft Corporation | Estimating Available Bandwidth With Multiple Overloading Streams |
US20080008095A1 (en) * | 2006-07-10 | 2008-01-10 | International Business Machines Corporation | Method for Distributed Traffic Shaping across a Cluster |
US20100208588A1 (en) * | 2005-08-19 | 2010-08-19 | Juniper Networks, Inc. | Shared shaping of network traffic |
US20130100955A1 (en) * | 2011-10-24 | 2013-04-25 | Qualcomm Incorporated | Technique for prioritizing traffic at a router |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7130268B2 (en) * | 2000-10-17 | 2006-10-31 | Saverio Mascolo | End-to-end bandwidth estimation for congestion control in packet switching networks |
US7826364B1 (en) * | 2006-02-09 | 2010-11-02 | Verizon Services Corp. | Dynamic service-aware flow control in packet networks |
US7835405B2 (en) | 2006-12-01 | 2010-11-16 | Ikanos Communications, Inc. | Multiplexing/demultiplexing on a shared interface |
JP4407700B2 (en) * | 2007-02-02 | 2010-02-03 | 日本電気株式会社 | Communication terminal, communication system, congestion control method, and congestion control program |
US20090034596A1 (en) | 2007-08-01 | 2009-02-05 | Acterna Llc | Ethernet Traffic Emulation Using Ramped Traffic Generation Techniques |
WO2009034450A2 (en) | 2007-09-13 | 2009-03-19 | Accedian Networks Inc. | System for testing ethernet paths and links without impacting non-test traffic |
DE102007050941A1 (en) | 2007-10-23 | 2009-04-30 | Phoenix Contact Gmbh & Co. Kg | Method for transmitting data in a communications network, transmitting and receiving device for a communications network |
US8730812B2 (en) | 2009-01-09 | 2014-05-20 | Brocade Communications Systems, Inc. | Hierarchical rate color marker |
US8139485B2 (en) | 2009-01-22 | 2012-03-20 | Ciena Corporation | Logical transport resource traffic management |
US8773993B2 (en) * | 2011-01-31 | 2014-07-08 | Apple Inc. | Adaptive bandwidth estimation |
US9025435B2 (en) * | 2012-10-18 | 2015-05-05 | Ciena Corporation | Optimized seam allocation in an ethernet ring network |
-
2012
- 2012-12-06 US US13/706,420 patent/US20140160945A1/en not_active Abandoned
-
2017
- 2017-07-06 US US15/643,027 patent/US10484291B2/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001091375A2 (en) * | 2000-05-19 | 2001-11-29 | Williams Communications Llc | Method and system for measuring one-way delay variation |
US20060215574A1 (en) * | 2005-03-24 | 2006-09-28 | Microsoft Corporation | Bandwidth estimation in broadband access networks |
US20100208588A1 (en) * | 2005-08-19 | 2010-08-19 | Juniper Networks, Inc. | Shared shaping of network traffic |
US20070081561A1 (en) * | 2005-10-11 | 2007-04-12 | International Business Machines Corporation | Single ended solution for estimation of bandwidth delay product |
US20070217448A1 (en) * | 2006-03-16 | 2007-09-20 | Microsoft Corporation | Estimating Available Bandwidth With Multiple Overloading Streams |
US20080008095A1 (en) * | 2006-07-10 | 2008-01-10 | International Business Machines Corporation | Method for Distributed Traffic Shaping across a Cluster |
US20130100955A1 (en) * | 2011-10-24 | 2013-04-25 | Qualcomm Incorporated | Technique for prioritizing traffic at a router |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150236962A1 (en) * | 2014-02-14 | 2015-08-20 | Exinda Networks PTY, Ltd. of Australia | Method and system for using dynamic bandwidth detection to drive quality of service control refinement |
US20190230001A1 (en) * | 2014-04-17 | 2019-07-25 | Accedian Networks Inc. | System and method for out-of-line real-time in-service performance measurement |
US10700941B2 (en) * | 2014-04-17 | 2020-06-30 | Accedian Networks Inc. | System and method for out-of-line real-time in-service performance measurement |
WO2016149866A1 (en) * | 2015-03-20 | 2016-09-29 | 华为技术有限公司 | Method and device for measuring available bandwidth in end to end path |
CN107210936A (en) * | 2015-03-20 | 2017-09-26 | 华为技术有限公司 | Availability bandwidth measurement method and device in end-to-end path |
US20180150432A1 (en) * | 2015-08-31 | 2018-05-31 | Huawei Technologies Co., Ltd. | Method and apparatus for measuring time samp unit of remote device |
US10503807B2 (en) * | 2015-08-31 | 2019-12-10 | Huawei Technologies Co., Ltd. | Method and apparatus for measuring time stamp unit of remote device |
Also Published As
Publication number | Publication date |
---|---|
US20170302583A1 (en) | 2017-10-19 |
US10484291B2 (en) | 2019-11-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10484291B2 (en) | Using bandwidth measurements to adjust CIR and EIR on a sub-rate link | |
Nasrallah et al. | Ultra-low latency (ULL) networks: The IEEE TSN and IETF DetNet standards and related 5G ULL research | |
WO2021083269A1 (en) | Network traffic rate limiting method and device, central control equipment and gateways | |
EP3172876B1 (en) | Ensuring dynamic traffic shaping fairness | |
US7756032B2 (en) | Method and apparatus for communicating data within measurement traffic | |
EP2742657B1 (en) | Dynamic bandwidth adjustment in packet transport network | |
US9451500B2 (en) | Conditional routing technique | |
US8009582B2 (en) | Method and apparatus for performance monitoring in a communications network | |
US10833974B2 (en) | Dynamic configuration of routing paths in a data network | |
US9288002B2 (en) | Method for monitoring and managing data networks | |
KR20130126816A (en) | Traffic management apparatus for controlling traffic congestion and method thereof | |
Olariu et al. | A QoS based call admission control and resource allocation mechanism for LTE femtocell deployment | |
US20230051166A1 (en) | Delay Sensitive Network Estimation System | |
Vdovin et al. | Network utilization optimizer for SD-WAN | |
JP4222422B2 (en) | Node, communication system and node program | |
WO2002033893A2 (en) | Method and apparatus for communicating data within measurement traffic | |
US20220271987A1 (en) | Methods and system for adaptive measurements applied to real time performance monitoring in a packet network | |
Nam et al. | Analysis of minimal backlogging-based available bandwidth estimation mechanism | |
Nurhaida et al. | INTERIOR GATEWAY PROTOCOL ROUTING PERFORMANCE COMPARISON OF THE VIRTUAL PRIVATE NETWORK BASED ON MULTI PROTOCOL LABEL SWITCHING AND DIRECT-LINK BACKUPSED ON MPLS AND DIRECT-LINK BACKUP | |
Ramadža et al. | Network performance monitoring within MPLS traffic engineering enabled networks | |
Suhaimy et al. | Analysis of MPLS-TP network for different applications | |
Kretschmer et al. | QoS-aware flow monitoring and event creation in heterogeneous MPLS-based wireless mesh networks supporting unidirectional links | |
Staehle et al. | Measuring one-way delay in wireless mesh networks: an experimental investigation | |
CA3149650C (en) | Methods and system for adaptive measurements applied to real time performance monitoring in a packet network | |
US11916777B2 (en) | Learning SLA violation probability from intelligent fine grained probing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: ACCEDIAN NETWORKS INC., CANADA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:ROBITAILLE, CLAUDE;REEL/FRAME:029416/0083 Effective date: 20121206 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |