WO2010114440A1 - An energy efficient ethernet network node and a method for use in an ethernet network node - Google Patents

An energy efficient ethernet network node and a method for use in an ethernet network node Download PDF

Info

Publication number
WO2010114440A1
WO2010114440A1 PCT/SE2009/050342 SE2009050342W WO2010114440A1 WO 2010114440 A1 WO2010114440 A1 WO 2010114440A1 SE 2009050342 W SE2009050342 W SE 2009050342W WO 2010114440 A1 WO2010114440 A1 WO 2010114440A1
Authority
WO
WIPO (PCT)
Prior art keywords
ethernet
link
single logical
network node
logical link
Prior art date
Application number
PCT/SE2009/050342
Other languages
French (fr)
Inventor
Tomas Thyni
Christian Gotare
Johan KÖLHI
Annikki Welin
Original Assignee
Telefonaktiebolaget L M Ericsson (Publ)
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Telefonaktiebolaget L M Ericsson (Publ) filed Critical Telefonaktiebolaget L M Ericsson (Publ)
Priority to PCT/SE2009/050342 priority Critical patent/WO2010114440A1/en
Publication of WO2010114440A1 publication Critical patent/WO2010114440A1/en

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/02—Details
    • H04L12/12—Arrangements for remote connection or disconnection of substations or of equipment thereof
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26—Power supply means, e.g. regulation thereof
    • G06F1/32—Means for saving power
    • G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26—Power supply means, e.g. regulation thereof
    • G06F1/32—Means for saving power
    • G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
    • G06F1/3234—Power saving characterised by the action undertaken
    • G06F1/325—Power saving in peripheral device
    • G06F1/3253—Power saving in bus
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/02—Details
    • H04L12/10—Current supply arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/2854—Wide area networks, e.g. public data networks
    • H04L12/2856—Access arrangements, e.g. Internet access
    • H04L12/2863—Arrangements for combining access network resources elements, e.g. channel bonding
    • H04L12/2867—Physical combinations
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/14—Multichannel or multilink protocols
    • 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
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
    • 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 present invention relates to the field of data communication, and in particular to an energy efficient Ethernet network node.
  • the present invention further relates to a method and a computer program product for use in an Ethernet network node.
  • Ethernet network is today a very common way of achieving data communication between network nodes in both local area networks (LANs) and wide area networks (WANs).
  • Ethernet communication is standardized by standard documentation, such as, for example, IEEE 802.3-2005, which defines how data transmissions are to be performed over the various layers of Ethernet network interfaces.
  • Ethernet network nodes may attach through the use of Ethernet network interfaces to a common medium that provides a path along which signals carrying data can travel in between the Ethernet network nodes.
  • This medium may, for example, be twisted pair copper cabling or fibre optic cabling.
  • a problem to which the present invention relates is the problem of providing Ethernet network nodes with reduced energy consumption.
  • a network node for data communication over a plurality of Ethernet links comprising at least a first and a second Ethernet link
  • said network node comprising: a data communication unit arranged to communicate over the at least first and second Ethernet link; and a control unit arranged to aggregate the at least first and second Ethernet link together in order to form a single logical link, and balance the data transfer load between the at least first and second Ethernet links comprised in the single logical link, characterized in that the control unit is further arranged to control the actual number of Ethernet links of said plurality of Ethernet links used in the single logical link at least partly based upon the actual data transfer load over the single logical link.
  • a method for use in a method for use in a network node for data communication over a plurality of Ethernet links comprising at least a first and a second Ethernet link comprising the steps of: aggregating the at least first and second Ethernet link together in order to form a single logical link; and balancing the data transfer loads between the at least first and second Ethernet link comprised in the single logical link, characterized by further comprising the steps of: monitoring the actual data transfer load over the single logical link; and controlling the actual number of Ethernet links of said plurality of Ethernet links used in the single logical link at least partly based upon the actual data transfer load over the single logical link.
  • control unit in a network node By having a control unit in a network node arranged to control the actual number of Ethernet links used in the single logical link based upon the actual data transfer load over the single logical link, allows the control unit in the network node to dynamically control the actual number of Ethernet links used in the single logical link in order to adapt the single logical link in view of the total actual data transfer load. This may reduce power- consumption during time periods where the full capacity of the single logical link is not utilized. This may be performed by setting removed Ethernet links in a standby idle mode, wherein various methods may be used to reduce the power consumption of the Ethernet link.
  • the control unit in the network node may be arranged to remove at least one Ethernet link from the actual number of Ethernet links of the plurality of Ethernet links used in the single logical link if the actual data transfer load over the single logical link is equal to or below a first predetermined threshold.
  • the first predetermined threshold may also be at least partly based on the total data transfer capacity of the actual number of Ethernet links of the plurality of Ethernet links remaining in the single logical link after the removal of the at least one Ethernet link. This allows dynamical threshold values to be used and for the control unit to perform a suitable control of the actual number of Ethernet links in the single logical link.
  • the control unit in the network node may also be arranged to re-distribute data traffic from the at least one Ethernet link to the actual number of Ethernet links of the plurality of Ethernet links remaining in the single logical link, for a predetermined time period prior to the removal of the at least one Ethernet link. This enables the control unit to perform pre- emptive and smooth transfers of data traffic from an Ethernet link that is about to be removed to the Ethernet links remaining in the single logical link without any data packet loss.
  • control unit in the network node may be further arranged to add at least one Ethernet link to the actual number of Ethernet links of the plurality of Ethernet links used in the single logical link if the actual data transfer load over the single logical link is equal to or above a second predetermined threshold.
  • the second predetermined threshold is at least partly based on the total data transfer capacity of the actual number of Ethernet links of the plurality of Ethernet links in the single logical link before the addition of the at least one Ethernet link. This further allows dynamical threshold values to be used and for the control unit to perform a suitable control of the actual number of Ethernet links in the single logical link.
  • control unit in the network node may further be arranged to put at least one removed Ethernet link in a standby idle mode, and also be arranged to add at least one Ethernet link to the actual number of Ethernet links of the plurality of Ethernet links used in the single logical link that is in a standby idle mode.
  • Another advantage of the above described invention is that it is particularly advantageous for network nodes comprising a data communication unit arranged to transmit data packets over an optical network, such as, a single mode optical fibre network or multimode optical fibre network.
  • an optical network such as, a single mode optical fibre network or multimode optical fibre network.
  • a computer program product for use in a network node as described above, which comprises computer readable code means, which when run in the control unit in the network node causes the control unit to perform the steps of: monitoring the actual data transfer load over a single logical link; and controlling the actual number of Ethernet links of a plurality of Ethernet links used in the single logical link based upon the actual data transfer load over the single logical link.
  • a system for data communication over a plurality of Ethernet links comprising at least a first and a second Ethernet link, said system comprising at least one network node as described above.
  • Fig. 1 shows network nodes in an Ethernet network according to prior art.
  • Fig. 2 illustrates Ethernet link aggregation in a network node according to the prior art.
  • Fig. 3-5 illustrates Ethernet link aggregation in a network node according to an exemplary embodiment of the invention.
  • Fig. 6 shows an example of data communication units and control units in network nodes according to an exemplary embodiment of the invention.
  • Fig. 7 shows a flowchart illustrating a method according to an exemplary embodiment of the invention.
  • Fig. 8 shows a flowchart illustrating a method according to another exemplary embodiment of the invention.
  • Fig. 1 shows an Ethernet network 100 according to prior art.
  • the Ethernet network 100 comprises a first Ethernet network node 101 and a second Ethernet network node 102.
  • the first and second Ethernet network node 101, 102 each comprises an Ethernet interface between which at least two Ethernet links 103, 104 can be established.
  • the medium supporting the Ethernet links 103, 104 over which the Ethernet interfaces of the first and second Ethernet network node 101, 102 are adapted to transmit and/or receive data may be e.g. twisted pair copper cabling or fibre optic cabling.
  • Data communication over the Ethernet links 103, 104, is performed by the first and second Ethernet network node 101, 102 in accordance with the Ethernet standard described in the standard documentation IEEE 802.3-2005, also sometimes referred to as IEEE 802.3ad. It should be noted that hereinafter, when reference is made to the Ethernet standard, the Ethernet standard according to and described in the standard documentation IEEE 802.3-2005 is intended.
  • Fig. 2 illustrates Ethernet link aggregation in a first Ethernet network node 101 according to the prior art.
  • Link aggregation describes using multiple physical network outputs, for example, network cables, ports, etc., in a network node 101 in parallel to increase the link speed beyond the limits of any one single physical network output in the network node 101.
  • Link aggregation may also increase the redundancy for higher availability.
  • Other terms that may be used to describe link aggregation includes "Ethernet trunk", "NIC teaming", “port channel”, “port teaming”, “port trunking”, “link bundling", “EtherChannel”, “Multi-Link Trunking (MLT)", “NIC bonding", “Network Fault Tolerance (NFT)” or the like.
  • Ethernet Link Aggregation Group (LAG) protocols allows a first Ethernet network node 101 to aggregate two or more Ethernet links 103, 104 so as to form an Ethernet Link Aggregation Group (LAG) 7.
  • An example of such an Ethernet Link Aggregation Protocol is the Link Aggregation Control Protocol (LACP) specified in the Ethernet standard, or the Cisco Port aggregation Protocol (PAgP).
  • LACP Link Aggregation Control Protocol
  • PAgP Cisco Port aggregation Protocol
  • the first Ethernet network node 101 is thus able to consider and treat the two or more Ethernet links 103, 104 comprised in the LAG 7 as a single logical Ethernet link between the Ethernet network node 101 and the Ethernet network node 102.
  • the Ethernet network nodes 101, 102 may also be logical nodes, wherein the link aggregation may be referred to as a Multi-Chassis Link Aggregation Group (MC-LAG).
  • M-LAG Multi-Chassis Link Aggregation Group
  • the actual data transfer load at any given time over the LAG 7 is not taken into consideration when using the LAG 7. It follows that for an exemplary setup comprising a LAG 7 including two or more Ethernet links between the two Ethernet network nodes 101, 102, all of the Ethernet links in the LAG 7 will be utilized for balancing the actual data transfer load even if the actual data transfer load at a given time is below the capacity of a single one of the Ethernet links in the LAG 7. Thus, the continuous, active operation of all of the transmitting/receiving components located in the first and second network nodes 1, 2 will lead to high power consumption in the first and second network nodes 1, 2.
  • LACP Link Aggregation Control Protocol
  • PAgP Cisco Port aggregation Protocol
  • the problem is addressed by having a control unit in a network node, which is arranged to aggregate Ethernet links into a single logical link and balance the data transfer load between the Ethernet links in the single logical link, being further arranged to control the actual number of Ethernet links used in the single logical link at least partly based upon the total actual data transfer load over the single logical link.
  • This allows a link aggregation network control protocol monitoring the total actual data transfer load over a single logical link to dynamically control (i.e. reduce or add to) the actual number of Ethernet links used in the single logical link in order to adapt the single logical link in dependence of the total actual data transfer load. This may reduce power-consumption during time periods where the full capacity of the single logical link is not utilized.
  • Fig. 3-5 shows a first network node 1 and a second network node 2 according to an exemplary embodiment of the invention.
  • the first and second network nodes 1, 2 may, for example, be comprised in a broadband multiplexer, a switch, a router, a server or the like.
  • the first and second network nodes 1, 2 may each comprise a control unit 3, 4 and a data communication unit 5, 6, respectively.
  • the control units 3, 4 may also be incorporated in the data communication units 5, 6, respectively.
  • the data communication units 5, 6 may be arranged to provide the first and second network nodes 1, 2 with an Ethernet network interface.
  • the data communication units 5, 6 may therefore also be referred to as Ethernet network cards, Ethernet network adapters, Ethernet network interface controllers (NICs), Ethernet network interface cards or chips, or Ethernet LAN adapters or the like.
  • the data communication units 5, 6 are arranged to establish and communicate over at least two Ethernet links 301, 302, 303, 304 in between the first and second network nodes 1, 2. It should be noted that the at least two Ethernet links 301, 302, 303, 304 may have any suitable Ethernet link speed or data transfer capacity, however, all of the at least two Ethernet links 301, 302, 303, 304 should have substantially the same or equal Ethernet link speed or data transfer capacity.
  • the data communication units 5, 6 may further comprise any suitable number of components which may provide suitable logic and circuitry that enable the data communication units 5, 6 to input/output streams of data through the at least two Ethernet links 301, 302, 303, 304 over a chosen medium.
  • Such data communication units 5, 6 are very common in the state of the art, and may be found in various different configurations and designs.
  • the control units 3, 4 are arranged to communicate with and control the data communication units 5, 6 respectively, and comprise logic for performing the functionality of the first and second network nodes 1, 2, respectively. This functionality may be implemented by means of a software or computer program.
  • the control units 3, 4 may also comprise storage means or a memory unit for storing the computer program and processing means or a processing unit, such as a microprocessor, for executing the computer program.
  • the storage means or memory unit is a readable storage medium.
  • the readable storage medium may also be separated from, but connected to the control units 3, 4.
  • the control units 3, 4 performs a certain function it is to be understood that the control units 3, 4 in the first and second network nodes 1, 2 uses the processing means to execute a certain part of the program which is stored in their storage means.
  • the control units 3, 4 and the data communication units 5, 6, respectively may comprise shared computational and storage capabilities, and may be provided as one physical unit, or alternatively as a plurality of logically interconnected units.
  • the control units 3, 4 and the data communication units 5, 6 may also be arranged to receive input and output data over a data communication interface 8, 9, such as, for example, a data communication bus or the like.
  • the control units 3, 4 and the data communication units 5, 6 may further be arranged to communicate over the data communication interface 8, 9 with other connected hosts, such as, for example, subsequent network nodes, computer systems or the like.
  • the control units 3, 4 in the network nodes 1, 2 are arranged to aggregate at least a first and a second Ethernet link 301, 302 of the at least two Ethernet links 301, 302, 303, 304 together in order to form a single logical link 7A-7C, or LAG, and balance the actual data transfer load between the at least first and second Ethernet link 301, 302 comprised in the LAG 7A-7C.
  • the control units 3, 4 are also arranged to dynamically control the actual number of the Ethernet links 301, 302, 303, 304 that are used in the LAG 7A-7C based upon the actual data transfer load over the LAG 7A-7C. An exemplary embodiment of this dynamic control of the control units 3, 4 is described in more detail below with reference to Figs. 3-5.
  • Ethernet links 301, 302, 303, 304 are comprised in an Ethernet Link
  • LAG Aggregation Group 7A. Therefore, the actual data transfer load from the network node 1 to the network node 2 is balanced between all four Ethernet links 301, 302, 303, 304, as is indicated by the percentages for each of the four Ethernet links 301, 302, 303, 304.
  • LACP Link Aggregation Control Protocol
  • PAgP Cisco Port aggregation Protocol
  • the control unit 3 may remove the Ethernet link 304 from the LAG 7 A.
  • the first data transfer load threshold may, for example, be based on the total data transfer capacity of the Ethernet links 301, 302, 303 remaining in the LAG 7A after the removal of the Ethernet link 304 from the LAG 7A. For example, at this stage, if the four Ethernet links 301, 302, 303, 304 in Figs. 3- 5 are four 10 Gigabit-Ethernet links, the first data transfer load threshold may be based on a value of 30 Gigabit/s.
  • control unit 3 may begin to remove the Ethernet link 304 from the LAG 7A.
  • the removed Ethernet link 304 may then be put in a standby idle mode by the control unit 3.
  • the control unit 3 may also re-distribute data traffic from the Ethernet link 304 to the remaining Ethernet links 301, 302, 303 in the LAG 7A for a predetermined time period prior to the removal of the Ethernet link 304 from the LAG 7 A. This allows the control unit 3 to pre-emptively transfer the data traffic of the Ethernet link 304 over to the remaining Ethernet links 301, 302, 303 in the LAG 7A without any data packet loss before the removal of the Ethernet link 304 from the LAG 7A.
  • the control unit 3 in the network node 1 has removed the Ethernet link 304 from the LAG 7A and put the Ethernet link 304 in a standby idle mode.
  • the control unit 3 is now arranged to balance the actual data transfer load from the network node 1 to the network node 2 between the Ethernet links 301, 302, 303 in the LAG 7B, as is indicated by the percentages for each of the Ethernet links 301, 302, 303.
  • the control unit 3 may add the Ethernet link 304 to the LAG 7B.
  • the second data transfer load threshold may, for example, be the same as the first data transfer load threshold or be based on the total data transfer capacity of the Ethernet links 301, 302, 303 in the LAG 7B before the addition of the Ethernet link 304.
  • the Ethernet link 304 may prior to being added to the
  • LAG 7B be in a standby idle mode or be inactive. In the latter case, the Ethernet link 304 would have to be re-established before being able to be added to the LAG 7B. After the addition of the Ethernet link 304, the LAG is configured according to LAG 7A as shown in Fig. 3.
  • the control unit 3 may remove the Ethernet link 303 from the LAG 7B.
  • the third data transfer load threshold may, for example, be based on the total data transfer capacity of the Ethernet links 301, 302 remaining in the LAG 7B after the removal of the Ethernet link 303 from the LAG 7B. For example, at this stage, if the four Ethernet links 301, 302, 303, 304 in Figs. 3-5 are four 10 Gigabit-Ethernet links, the third data transfer load threshold may be based on a value of 20 Gigabit/s.
  • control unit 3 may begin to remove the Ethernet link 303 from the LAG 7B.
  • the removed Ethernet link 303 may then be put in a standby idle mode by the control unit 3.
  • the control unit 3 in the network node 1 has removed the Ethernet link 303 from the LAG 7B and put the Ethernet link 303 in a standby idle mode.
  • the control unit 3 is then arranged to balance the actual data transfer load from the network node 1 to the network node 2 between the Ethernet links 301, 302 in the LAG 7C, as is indicated by the percentages for each of the Ethernet links 301, 302.
  • the control unit 3 may add the Ethernet link 303 to the LAG 7C.
  • the fourth data transfer load threshold may, for example, be the same as the third data transfer load threshold or be based on the total data transfer capacity of the Ethernet links 301, 302 in the LAG 7C before the addition of the Ethernet link 303.
  • the Ethernet link 303 may prior to being added to the LAG 7C be in a standby idle mode or be inactive. In the latter case, the Ethernet link 303 would have to be re-established before being able to be added to the LAG 7C.
  • the LAG is configured according to LAG 7B as shown in Fig. 4.
  • control units 3, 4 in the network nodes 1, 2 may be arranged to monitor the actual data transfer load over the single logical link, or LAG 7A, 7B, 7C, by using a statistic monitoring tool, such as, for example, Remote Monitoring (RMON).
  • RMON Remote Monitoring
  • the control unit 3, 4 may compile a value of the actual data transfer load, or LAG usage, over a predetermined period of time based on the RMON statistics. For example, if the four Ethernet links 301, 302, 303, 304 in Fig.
  • the fourth threshold may be configured such that if the actual data transfer load, or LAG usage, based on the RMON statistics over a 5 minute period is above 75 % of the capacity of the active Ethernet links 301, 302 (i.e. 37.5 % of the total logical capacity of the LAG 7A, 7B, 7C) the fourth threshold is exceeded and the Ethernet link 303 is added to the LAG 7C.
  • the first and third threshold may be configured such that the Ethernet link 304 and the Ethernet link 303 are removed when the actual data transfer load over the LAG is significantly below 30 Gigabits/s or 20 Gigabits/s, respectively. This may, for example, ensure that the removal of an Ethernet link from the LAG 7A, 7B, 7C is not performed unnecessarily.
  • a corresponding configuration may also apply in the case of an addition of an Ethernet link to the LAG 7A, 7B, 7C.
  • the number of the at least two Ethernet links exceeds the four Ethernet links 301, 302, 303, 304 of the example shown in Fig. 3-5, it may easily be understood that the number of data transfer load thresholds may be increased accordingly.
  • a dynamical control of the actual number of Ethernet links used in the LAG 7A, 7B, 7C may be achieved by the control units 3, 4.
  • Fig. 6 shows data communication units 5, 6 in the network nodes 1, 2 in more detail according to an exemplary embodiment of the invention.
  • the inventive features described herein are applicable to any design, configuration and/or amount of hardware components used in the data communication units 5, 6, any detailed description of the large amount of variations of these components, their designs and/or collective configurations in the data communication units 5, 6 are considered superfluous.
  • the components 5 A, 5B, 6A, 6B in the data communication units 5, 6 are in Fig. 6 divided into actual input/output medium interface components 5A, 6A and underlying supporting electrical/optical components 5B, 6B.
  • the chosen medium for the Ethernet link 304 is an optical medium
  • the input/output interface components 5 A, 6A may comprise an optical transmitter/receiver, optical laser transmitter/receiver or the like.
  • the input/output interface components 5 A, 6 A may comprise the transmitter/receiver or the like.
  • the data communication units 5, 6 may be arranged to communicate over, for example, a copper wire twisted pair medium, an optical medium (such as a single or multimode fibre network) or the like.
  • a copper wire twisted pair medium there are several forms of Ethernet standards, such as, for example, 1 OBASE-T, 100BASE-TX, and IOOOB ASE-T, etc.
  • Ethernet standards such as, for example, 1 OBASE-T, 100BASE-TX, and IOOOB ASE-T, etc.
  • optical fast-Ethernet standards such as, 100BASE-FX, 10OB ASE-SX and IOOOB ASE-LX, etc.
  • the data communication units 5, 6 may be arranged to communicate using any one of these standards or the like.
  • the medium supporting the Ethernet link 304 comprises fibre optic cabling
  • the power consumption of the optical and/or electronic components and circuitry will be larger than compared to, for example, twisted pair copper cabling.
  • the fibre optic cabling has some advantages over twisted pair copper cabling in that it provides, for example, improved performance, higher data rates, and enables large distances between the network node 1 and the network node 2, the fibre optic cabling also requires optical and/or electrical components and circuitry which conventionally consumes more electricity than, for example, traditional twisted pair copper wire solutions.
  • Some existing modes of operation for use in Ethernet network nodes are directed towards reducing the energy consumption of the optical and/or electronic components and circuitry when an Ethernet network node 1, 2 at one end of the Ethernet link 304 is disconnected or powered off.
  • Fig. 7 shows a flowchart illustrating a method according to an exemplary embodiment of the invention. It describes a general method for use in a network node as described above.
  • step S71 the control unit in the network node monitors the actual data transfer load over a single logical link (LAG).
  • the control unit may monitor the actual data transfer load in view of different thresholds and detect when the actual data transfer load either exceeds or surpasses a particular threshold. The monitoring may, for example, be based on RMON statistics and be made over different periods of time. If a particular threshold in exceeded or surpassed by the actual data transfer load the control unit may control the actual number of Ethernet links in the single logical link in step S72.
  • Fig. 8 shows a flowchart illustrating a further method according to an exemplary embodiment of the invention.
  • Step S81 is identical to step S71 according to the method shown in Fig. 7, and the steps S82-S85 describes a way to perform the control in step S72 according to the method shown in Fig. 7.
  • step S82 if the actual data transfer load, which is monitored by the control unit according to step S81, is equal to or lower than a first threshold, the control unit may remove an Ethernet link from the single logical link in step S83. The control unit may then also update the first threshold according to the new number of Ethernet links in the single logical link, and continue monitoring the actual data transfer load in step S81.
  • step S84 if the actual data transfer load is equal to or higher than a second threshold, the control unit may, if possible, add an Ethernet link to the single logical link in step S85. The control unit may then also update the second threshold according to the new number of Ethernet links in the single logical link, and continue monitoring the actual data transfer load in step S81. If none of the conditions as indicated in step S 82 and S 84 are fulfilled the control unit continues to monitor the actual data transfer load in step S81.
  • Another advantage of the invention is that it can be added as an extension of the existing Standard and products. The invention may directly be implemented in existing products without requiring an update of the hardware.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Security & Cryptography (AREA)
  • Small-Scale Networks (AREA)

Abstract

The present invention relates to an network node for data communication over a plurality of Ethernet links comprising at least a first and a second Ethernet link, the network node comprising: a data communication unit arranged to communicate over the at least first and second Ethernet link; and a control unit arranged to aggregate the at least first and second Ethernet link together in order to form a single logical link, and balance the data transfer load between the at least first and second Ethernet links comprised in the single logical link. The network node is characterized in that the control unit is further arranged to control the actual number of Ethernet links of the plurality of Ethernet links used in the single logical link based upon the actual data transfer load over the single logical link. The present invention further relates to a method and a computer program product for use in such as network node, and a system comprising such a network node.

Description

An energy efficient Ethernet network node and a method for use in an Ethernet network node
Technical field
The present invention relates to the field of data communication, and in particular to an energy efficient Ethernet network node. The present invention further relates to a method and a computer program product for use in an Ethernet network node.
Background
An Ethernet network is today a very common way of achieving data communication between network nodes in both local area networks (LANs) and wide area networks (WANs). Ethernet communication is standardized by standard documentation, such as, for example, IEEE 802.3-2005, which defines how data transmissions are to be performed over the various layers of Ethernet network interfaces.
Ethernet network nodes may attach through the use of Ethernet network interfaces to a common medium that provides a path along which signals carrying data can travel in between the Ethernet network nodes. This medium may, for example, be twisted pair copper cabling or fibre optic cabling.
However, with an increasing number Ethernet nodes connected in data communication networks and thus an increase in power-consuming Ethernet interface electronics, along with an increased cost-effective and environmental interest in saving energy, there is a need to provide more energy efficient Ethernet network nodes.
Summary
A problem to which the present invention relates is the problem of providing Ethernet network nodes with reduced energy consumption.
This problem is addressed by a network node for data communication over a plurality of Ethernet links comprising at least a first and a second Ethernet link, said network node comprising: a data communication unit arranged to communicate over the at least first and second Ethernet link; and a control unit arranged to aggregate the at least first and second Ethernet link together in order to form a single logical link, and balance the data transfer load between the at least first and second Ethernet links comprised in the single logical link, characterized in that the control unit is further arranged to control the actual number of Ethernet links of said plurality of Ethernet links used in the single logical link at least partly based upon the actual data transfer load over the single logical link.
The problem is also addressed by a method for use in a method for use in a network node for data communication over a plurality of Ethernet links comprising at least a first and a second Ethernet link, said method comprising the steps of: aggregating the at least first and second Ethernet link together in order to form a single logical link; and balancing the data transfer loads between the at least first and second Ethernet link comprised in the single logical link, characterized by further comprising the steps of: monitoring the actual data transfer load over the single logical link; and controlling the actual number of Ethernet links of said plurality of Ethernet links used in the single logical link at least partly based upon the actual data transfer load over the single logical link.
By having a control unit in a network node arranged to control the actual number of Ethernet links used in the single logical link based upon the actual data transfer load over the single logical link, allows the control unit in the network node to dynamically control the actual number of Ethernet links used in the single logical link in order to adapt the single logical link in view of the total actual data transfer load. This may reduce power- consumption during time periods where the full capacity of the single logical link is not utilized. This may be performed by setting removed Ethernet links in a standby idle mode, wherein various methods may be used to reduce the power consumption of the Ethernet link.
According to one aspect of the invention, the control unit in the network node may be arranged to remove at least one Ethernet link from the actual number of Ethernet links of the plurality of Ethernet links used in the single logical link if the actual data transfer load over the single logical link is equal to or below a first predetermined threshold. This allows the control unit to automatically detect when an Ethernet link should be removed from the single logical link. The first predetermined threshold may also be at least partly based on the total data transfer capacity of the actual number of Ethernet links of the plurality of Ethernet links remaining in the single logical link after the removal of the at least one Ethernet link. This allows dynamical threshold values to be used and for the control unit to perform a suitable control of the actual number of Ethernet links in the single logical link.
The control unit in the network node may also be arranged to re-distribute data traffic from the at least one Ethernet link to the actual number of Ethernet links of the plurality of Ethernet links remaining in the single logical link, for a predetermined time period prior to the removal of the at least one Ethernet link. This enables the control unit to perform pre- emptive and smooth transfers of data traffic from an Ethernet link that is about to be removed to the Ethernet links remaining in the single logical link without any data packet loss.
According to another aspect of the invention, the control unit in the network node may be further arranged to add at least one Ethernet link to the actual number of Ethernet links of the plurality of Ethernet links used in the single logical link if the actual data transfer load over the single logical link is equal to or above a second predetermined threshold. This allows the control unit to automatically detect when an Ethernet link should be added to the single logical link. The second predetermined threshold is at least partly based on the total data transfer capacity of the actual number of Ethernet links of the plurality of Ethernet links in the single logical link before the addition of the at least one Ethernet link. This further allows dynamical threshold values to be used and for the control unit to perform a suitable control of the actual number of Ethernet links in the single logical link.
According to a further aspect of the invention, the control unit in the network node may further be arranged to put at least one removed Ethernet link in a standby idle mode, and also be arranged to add at least one Ethernet link to the actual number of Ethernet links of the plurality of Ethernet links used in the single logical link that is in a standby idle mode.
Another advantage of the above described invention is that it is particularly advantageous for network nodes comprising a data communication unit arranged to transmit data packets over an optical network, such as, a single mode optical fibre network or multimode optical fibre network. This is because the optical and/or electronic components and circuitry in such data communication units, such as, optical transmitters or lasers, are often more energy consuming than the electronic components and circuitry used in, for example, a twisted copper pair network.
The problem stated above is further addressed by a computer program product for use in a network node as described above, which comprises computer readable code means, which when run in the control unit in the network node causes the control unit to perform the steps of: monitoring the actual data transfer load over a single logical link; and controlling the actual number of Ethernet links of a plurality of Ethernet links used in the single logical link based upon the actual data transfer load over the single logical link.
The problem stated above is further addressed by a system for data communication over a plurality of Ethernet links comprising at least a first and a second Ethernet link, said system comprising at least one network node as described above.
Further advantageous embodiments of the method and computer program product for use in a network node are set forth in the dependent claims and correspond to advantageous embodiments already set forth with reference to the previously mentioned network node.
Brief description of the drawings The objects, advantages and effects as well as features of the invention will be more readily understood from the following detailed description of exemplary embodiments of the invention when read together with the accompanying drawings, in which:
Fig. 1 shows network nodes in an Ethernet network according to prior art.
Fig. 2 illustrates Ethernet link aggregation in a network node according to the prior art.
Fig. 3-5 illustrates Ethernet link aggregation in a network node according to an exemplary embodiment of the invention.
Fig. 6 shows an example of data communication units and control units in network nodes according to an exemplary embodiment of the invention. Fig. 7 shows a flowchart illustrating a method according to an exemplary embodiment of the invention.
Fig. 8 shows a flowchart illustrating a method according to another exemplary embodiment of the invention.
Detailed description
Fig. 1 shows an Ethernet network 100 according to prior art. The Ethernet network 100 comprises a first Ethernet network node 101 and a second Ethernet network node 102. The first and second Ethernet network node 101, 102 each comprises an Ethernet interface between which at least two Ethernet links 103, 104 can be established. The medium supporting the Ethernet links 103, 104 over which the Ethernet interfaces of the first and second Ethernet network node 101, 102 are adapted to transmit and/or receive data, may be e.g. twisted pair copper cabling or fibre optic cabling. Data communication over the Ethernet links 103, 104, is performed by the first and second Ethernet network node 101, 102 in accordance with the Ethernet standard described in the standard documentation IEEE 802.3-2005, also sometimes referred to as IEEE 802.3ad. It should be noted that hereinafter, when reference is made to the Ethernet standard, the Ethernet standard according to and described in the standard documentation IEEE 802.3-2005 is intended.
Fig. 2 illustrates Ethernet link aggregation in a first Ethernet network node 101 according to the prior art. Link aggregation describes using multiple physical network outputs, for example, network cables, ports, etc., in a network node 101 in parallel to increase the link speed beyond the limits of any one single physical network output in the network node 101. Link aggregation may also increase the redundancy for higher availability. Other terms that may be used to describe link aggregation includes "Ethernet trunk", "NIC teaming", "port channel", "port teaming", "port trunking", "link bundling", "EtherChannel", "Multi-Link Trunking (MLT)", "NIC bonding", "Network Fault Tolerance (NFT)" or the like.
Existing Ethernet Link Aggregation Group (LAG) protocols allows a first Ethernet network node 101 to aggregate two or more Ethernet links 103, 104 so as to form an Ethernet Link Aggregation Group (LAG) 7. An example of such an Ethernet Link Aggregation Protocol is the Link Aggregation Control Protocol (LACP) specified in the Ethernet standard, or the Cisco Port aggregation Protocol (PAgP). The first Ethernet network node 101 is thus able to consider and treat the two or more Ethernet links 103, 104 comprised in the LAG 7 as a single logical Ethernet link between the Ethernet network node 101 and the Ethernet network node 102. The Ethernet network nodes 101, 102 may also be logical nodes, wherein the link aggregation may be referred to as a Multi-Chassis Link Aggregation Group (MC-LAG).
However, a problem experienced in such conventional Ethernet network nodes and in such Ethernet networks as presented above is to provide more energy efficient Ethernet network nodes with reduced power consumption.
In existing Ethernet LAG protocols, such as, the Link Aggregation Control Protocol (LACP) specified in the Ethernet standard or the Cisco Port aggregation Protocol (PAgP), the actual data transfer load at any given time over the LAG 7 is not taken into consideration when using the LAG 7. It follows that for an exemplary setup comprising a LAG 7 including two or more Ethernet links between the two Ethernet network nodes 101, 102, all of the Ethernet links in the LAG 7 will be utilized for balancing the actual data transfer load even if the actual data transfer load at a given time is below the capacity of a single one of the Ethernet links in the LAG 7. Thus, the continuous, active operation of all of the transmitting/receiving components located in the first and second network nodes 1, 2 will lead to high power consumption in the first and second network nodes 1, 2.
According to the inventive features of the invention, the problem is addressed by having a control unit in a network node, which is arranged to aggregate Ethernet links into a single logical link and balance the data transfer load between the Ethernet links in the single logical link, being further arranged to control the actual number of Ethernet links used in the single logical link at least partly based upon the total actual data transfer load over the single logical link. This allows a link aggregation network control protocol monitoring the total actual data transfer load over a single logical link to dynamically control (i.e. reduce or add to) the actual number of Ethernet links used in the single logical link in order to adapt the single logical link in dependence of the total actual data transfer load. This may reduce power-consumption during time periods where the full capacity of the single logical link is not utilized. An advantageous exemplary embodiment of the invention is described in more detail below with reference to Figs. 3-8.
Fig. 3-5 shows a first network node 1 and a second network node 2 according to an exemplary embodiment of the invention. The first and second network nodes 1, 2 may, for example, be comprised in a broadband multiplexer, a switch, a router, a server or the like. The first and second network nodes 1, 2 may each comprise a control unit 3, 4 and a data communication unit 5, 6, respectively. The control units 3, 4 may also be incorporated in the data communication units 5, 6, respectively.
The data communication units 5, 6 may be arranged to provide the first and second network nodes 1, 2 with an Ethernet network interface. The data communication units 5, 6 may therefore also be referred to as Ethernet network cards, Ethernet network adapters, Ethernet network interface controllers (NICs), Ethernet network interface cards or chips, or Ethernet LAN adapters or the like. The data communication units 5, 6 are arranged to establish and communicate over at least two Ethernet links 301, 302, 303, 304 in between the first and second network nodes 1, 2. It should be noted that the at least two Ethernet links 301, 302, 303, 304 may have any suitable Ethernet link speed or data transfer capacity, however, all of the at least two Ethernet links 301, 302, 303, 304 should have substantially the same or equal Ethernet link speed or data transfer capacity. The data communication units 5, 6 may further comprise any suitable number of components which may provide suitable logic and circuitry that enable the data communication units 5, 6 to input/output streams of data through the at least two Ethernet links 301, 302, 303, 304 over a chosen medium. Such data communication units 5, 6 are very common in the state of the art, and may be found in various different configurations and designs.
The control units 3, 4 are arranged to communicate with and control the data communication units 5, 6 respectively, and comprise logic for performing the functionality of the first and second network nodes 1, 2, respectively. This functionality may be implemented by means of a software or computer program. The control units 3, 4 may also comprise storage means or a memory unit for storing the computer program and processing means or a processing unit, such as a microprocessor, for executing the computer program.
The storage means or memory unit is a readable storage medium. The readable storage medium may also be separated from, but connected to the control units 3, 4. When, in the following, it is described that the control units 3, 4 performs a certain function it is to be understood that the control units 3, 4 in the first and second network nodes 1, 2 uses the processing means to execute a certain part of the program which is stored in their storage means. It should also be noted that the control units 3, 4 and the data communication units 5, 6, respectively, may comprise shared computational and storage capabilities, and may be provided as one physical unit, or alternatively as a plurality of logically interconnected units. The control units 3, 4 and the data communication units 5, 6 may also be arranged to receive input and output data over a data communication interface 8, 9, such as, for example, a data communication bus or the like. The control units 3, 4 and the data communication units 5, 6 may further be arranged to communicate over the data communication interface 8, 9 with other connected hosts, such as, for example, subsequent network nodes, computer systems or the like.
The control units 3, 4 in the network nodes 1, 2 are arranged to aggregate at least a first and a second Ethernet link 301, 302 of the at least two Ethernet links 301, 302, 303, 304 together in order to form a single logical link 7A-7C, or LAG, and balance the actual data transfer load between the at least first and second Ethernet link 301, 302 comprised in the LAG 7A-7C. The control units 3, 4 are also arranged to dynamically control the actual number of the Ethernet links 301, 302, 303, 304 that are used in the LAG 7A-7C based upon the actual data transfer load over the LAG 7A-7C. An exemplary embodiment of this dynamic control of the control units 3, 4 is described in more detail below with reference to Figs. 3-5.
In Fig. 3, four Ethernet links 301, 302, 303, 304 are comprised in an Ethernet Link
Aggregation Group (LAG) 7A. Therefore, the actual data transfer load from the network node 1 to the network node 2 is balanced between all four Ethernet links 301, 302, 303, 304, as is indicated by the percentages for each of the four Ethernet links 301, 302, 303, 304. For existing Ethernet LAG protocols, such as, the Link Aggregation Control Protocol (LACP) specified in the Ethernet standard or the Cisco Port aggregation Protocol (PAgP), this will be true for all actual data transfer loads at any given time, even if the actual data transfer load is below the capacity of a single one of the four Ethernet links 301, 302, 303, 304 in the LAG 7A. However, according to the invention, if the actual data transfer load over the LAG 7 A drops below a first data transfer load threshold comprised in the control unit 3, the control unit 3 may remove the Ethernet link 304 from the LAG 7 A. The first data transfer load threshold may, for example, be based on the total data transfer capacity of the Ethernet links 301, 302, 303 remaining in the LAG 7A after the removal of the Ethernet link 304 from the LAG 7A. For example, at this stage, if the four Ethernet links 301, 302, 303, 304 in Figs. 3- 5 are four 10 Gigabit-Ethernet links, the first data transfer load threshold may be based on a value of 30 Gigabit/s. Thus, if the actual data transfer load over the LAG 7A drops below 30 Gigabit/s, the control unit 3 may begin to remove the Ethernet link 304 from the LAG 7A. The removed Ethernet link 304 may then be put in a standby idle mode by the control unit 3.
The control unit 3 may also re-distribute data traffic from the Ethernet link 304 to the remaining Ethernet links 301, 302, 303 in the LAG 7A for a predetermined time period prior to the removal of the Ethernet link 304 from the LAG 7 A. This allows the control unit 3 to pre-emptively transfer the data traffic of the Ethernet link 304 over to the remaining Ethernet links 301, 302, 303 in the LAG 7A without any data packet loss before the removal of the Ethernet link 304 from the LAG 7A.
In Fig. 4, the control unit 3 in the network node 1 has removed the Ethernet link 304 from the LAG 7A and put the Ethernet link 304 in a standby idle mode. The control unit 3 is now arranged to balance the actual data transfer load from the network node 1 to the network node 2 between the Ethernet links 301, 302, 303 in the LAG 7B, as is indicated by the percentages for each of the Ethernet links 301, 302, 303.
At this stage, if the actual data transfer load over the LAG 7B should climb such that it is equal to or above a second data transfer load threshold comprised in the control unit 3, the control unit 3 may add the Ethernet link 304 to the LAG 7B. The second data transfer load threshold may, for example, be the same as the first data transfer load threshold or be based on the total data transfer capacity of the Ethernet links 301, 302, 303 in the LAG 7B before the addition of the Ethernet link 304. The Ethernet link 304 may prior to being added to the
LAG 7B be in a standby idle mode or be inactive. In the latter case, the Ethernet link 304 would have to be re-established before being able to be added to the LAG 7B. After the addition of the Ethernet link 304, the LAG is configured according to LAG 7A as shown in Fig. 3.
However, if the actual data transfer load over the LAG 7B in Fig. 4 instead drops below a third data transfer load threshold comprised in the control unit 3, the control unit 3 may remove the Ethernet link 303 from the LAG 7B. The third data transfer load threshold may, for example, be based on the total data transfer capacity of the Ethernet links 301, 302 remaining in the LAG 7B after the removal of the Ethernet link 303 from the LAG 7B. For example, at this stage, if the four Ethernet links 301, 302, 303, 304 in Figs. 3-5 are four 10 Gigabit-Ethernet links, the third data transfer load threshold may be based on a value of 20 Gigabit/s. Thus, if the actual data transfer load over the LAG 7B drops below 20 Gigabit/s, the control unit 3 may begin to remove the Ethernet link 303 from the LAG 7B. The removed Ethernet link 303 may then be put in a standby idle mode by the control unit 3.
In Fig. 5, the control unit 3 in the network node 1 has removed the Ethernet link 303 from the LAG 7B and put the Ethernet link 303 in a standby idle mode. The control unit 3 is then arranged to balance the actual data transfer load from the network node 1 to the network node 2 between the Ethernet links 301, 302 in the LAG 7C, as is indicated by the percentages for each of the Ethernet links 301, 302.
At this stage, if the actual data transfer load over the LAG 7C should climb such that it is equal to or above a fourth data transfer load threshold comprised in the control unit 3, the control unit 3 may add the Ethernet link 303 to the LAG 7C. The fourth data transfer load threshold may, for example, be the same as the third data transfer load threshold or be based on the total data transfer capacity of the Ethernet links 301, 302 in the LAG 7C before the addition of the Ethernet link 303. The Ethernet link 303 may prior to being added to the LAG 7C be in a standby idle mode or be inactive. In the latter case, the Ethernet link 303 would have to be re-established before being able to be added to the LAG 7C. After the addition of the Ethernet link 303, the LAG is configured according to LAG 7B as shown in Fig. 4.
Also, the control units 3, 4 in the network nodes 1, 2 may be arranged to monitor the actual data transfer load over the single logical link, or LAG 7A, 7B, 7C, by using a statistic monitoring tool, such as, for example, Remote Monitoring (RMON). The control unit 3, 4 may compile a value of the actual data transfer load, or LAG usage, over a predetermined period of time based on the RMON statistics. For example, if the four Ethernet links 301, 302, 303, 304 in Fig. 5 are four 10 Gigabit-Ethernet links, wherein the Ethernet links 301, 302 are active and the Ethernet links 303, 304 are in a standby idle mode, the fourth threshold may be configured such that if the actual data transfer load, or LAG usage, based on the RMON statistics over a 5 minute period is above 75 % of the capacity of the active Ethernet links 301, 302 (i.e. 37.5 % of the total logical capacity of the LAG 7A, 7B, 7C) the fourth threshold is exceeded and the Ethernet link 303 is added to the LAG 7C.
It should be understood that solely the principle guiding the choice of the first, second, third and fourth data transfer load threshold is described above, and that it may be the subject of many variations, various design constraints, application specific implementations or the like. For example, if the four Ethernet links 301, 302, 303, 304 in Figs. 3-5 are four 10 Gigabit- Ethernet links, the first and third threshold may be configured such that the Ethernet link 304 and the Ethernet link 303 are removed when the actual data transfer load over the LAG is significantly below 30 Gigabits/s or 20 Gigabits/s, respectively. This may, for example, ensure that the removal of an Ethernet link from the LAG 7A, 7B, 7C is not performed unnecessarily. A corresponding configuration may also apply in the case of an addition of an Ethernet link to the LAG 7A, 7B, 7C. In case the number of the at least two Ethernet links exceeds the four Ethernet links 301, 302, 303, 304 of the example shown in Fig. 3-5, it may easily be understood that the number of data transfer load thresholds may be increased accordingly. Thus, a dynamical control of the actual number of Ethernet links used in the LAG 7A, 7B, 7C may be achieved by the control units 3, 4.
Fig. 6 shows data communication units 5, 6 in the network nodes 1, 2 in more detail according to an exemplary embodiment of the invention.
Since the inventive features described herein are applicable to any design, configuration and/or amount of hardware components used in the data communication units 5, 6, any detailed description of the large amount of variations of these components, their designs and/or collective configurations in the data communication units 5, 6 are considered superfluous. However, the components 5 A, 5B, 6A, 6B in the data communication units 5, 6 are in Fig. 6 divided into actual input/output medium interface components 5A, 6A and underlying supporting electrical/optical components 5B, 6B. As an example, if the chosen medium for the Ethernet link 304 is an optical medium, the input/output interface components 5 A, 6A may comprise an optical transmitter/receiver, optical laser transmitter/receiver or the like. According to another example, if the chosen medium for the Ethernet link 304 is a twisted pair medium, the input/output interface components 5 A, 6 A may comprise the transmitter/receiver or the like.
The data communication units 5, 6 may be arranged to communicate over, for example, a copper wire twisted pair medium, an optical medium (such as a single or multimode fibre network) or the like. For transmission over copper wire twisted pair mediums, there are several forms of Ethernet standards, such as, for example, 1 OBASE-T, 100BASE-TX, and IOOOB ASE-T, etc. For transmission over optical mediums, there are several forms of optical fast-Ethernet standards, such as, 100BASE-FX, 10OB ASE-SX and IOOOB ASE-LX, etc. The data communication units 5, 6 may be arranged to communicate using any one of these standards or the like. Furthermore, if the medium supporting the Ethernet link 304 comprises fibre optic cabling, the power consumption of the optical and/or electronic components and circuitry will be larger than compared to, for example, twisted pair copper cabling. This is because, although the fibre optic cabling has some advantages over twisted pair copper cabling in that it provides, for example, improved performance, higher data rates, and enables large distances between the network node 1 and the network node 2, the fibre optic cabling also requires optical and/or electrical components and circuitry which conventionally consumes more electricity than, for example, traditional twisted pair copper wire solutions.
Some existing modes of operation for use in Ethernet network nodes, commonly referred to as low power link state functions, green Ethernet or the like, are directed towards reducing the energy consumption of the optical and/or electronic components and circuitry when an Ethernet network node 1, 2 at one end of the Ethernet link 304 is disconnected or powered off. There are also energy-saving solutions directed towards the energy consumption of the actual input/output medium interface components 5 A, 6A and/or the underlying supporting electrical/optical components 5B, 6B in network nodes having active Ethernet links. Any one of these may be used by the control units 3, 4, for example, as standby idle modes for the Ethernet links that are not active in the LAG 7 A, 7B, 7C, and may thus aid in reducing the power consumption of the network nodes 1, 2.
Fig. 7 shows a flowchart illustrating a method according to an exemplary embodiment of the invention. It describes a general method for use in a network node as described above.
In step S71, the control unit in the network node monitors the actual data transfer load over a single logical link (LAG). The control unit may monitor the actual data transfer load in view of different thresholds and detect when the actual data transfer load either exceeds or surpasses a particular threshold. The monitoring may, for example, be based on RMON statistics and be made over different periods of time. If a particular threshold in exceeded or surpassed by the actual data transfer load the control unit may control the actual number of Ethernet links in the single logical link in step S72.
Fig. 8 shows a flowchart illustrating a further method according to an exemplary embodiment of the invention. Step S81 is identical to step S71 according to the method shown in Fig. 7, and the steps S82-S85 describes a way to perform the control in step S72 according to the method shown in Fig. 7.
In step S82, if the actual data transfer load, which is monitored by the control unit according to step S81, is equal to or lower than a first threshold, the control unit may remove an Ethernet link from the single logical link in step S83. The control unit may then also update the first threshold according to the new number of Ethernet links in the single logical link, and continue monitoring the actual data transfer load in step S81.
In step S84, if the actual data transfer load is equal to or higher than a second threshold, the control unit may, if possible, add an Ethernet link to the single logical link in step S85. The control unit may then also update the second threshold according to the new number of Ethernet links in the single logical link, and continue monitoring the actual data transfer load in step S81. If none of the conditions as indicated in step S 82 and S 84 are fulfilled the control unit continues to monitor the actual data transfer load in step S81. Another advantage of the invention is that it can be added as an extension of the existing Standard and products. The invention may directly be implemented in existing products without requiring an update of the hardware.
The description above is of the best mode presently contemplated for practising the present invention. The description is not intended to be taken in a limiting sense, but is made merely for the purpose of describing the general principles of the invention. The scope of the present invention should only be ascertained with reference to the issued claims.

Claims

1. A network node (1, 2) for data communication over a plurality of Ethernet links (301, 302, 303, 304) comprising at least a first and a second Ethernet link, said network node (1, 2) comprising: a data communication unit (5, 6) arranged to communicate over the at least first and second Ethernet link; and a control unit (3, 4) arranged to aggregate the at least first and second Ethernet link together in order to form a single logical link (7, 7A, 7B, 7C), and balance the data transfer loads between the at least first and second Ethernet links comprised in the single logical link (7, 7A, 7B, 7C),
characterized in that
the control unit (3, 4) is further arranged to control the actual number of
Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) used in the single logical link (7, 7A, 7B, 7C) at least partly based upon the actual data transfer load over the single logical link (7, 7A, 7B, 7C).
2. A network node (1, 2) according to claim 1, wherein the control unit (3, 4) is arranged to remove at least one Ethernet link from the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) used in the single logical link (7, 7A, 7B, 7C) if the actual data transfer load over the single logical link (7, 7 A, 7B, 7C) is equal to or below a first predetermined threshold.
3. A network node (1, 2) according to claim 2, wherein the first predetermined threshold is at least partly based on the total data transfer capacity of the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) remaining in the single logical link (7, 7A, 7B, 7C) after the removal of said at least one Ethernet link.
4. A network node (1, 2) according to claim 2 or 3, wherein the control unit (3, 4) is arranged to re-distribute data traffic from the at least one Ethernet link to the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) remaining in the single logical link (7, 7 A, 7B, 7C), for a predetermined time period prior to the removal of said at least one Ethernet link.
5. A network node (1, 2) according to any one of the claims 2-4, wherein the control unit (3, 4) is further arranged to put said at least one removed Ethernet link in a standby idle mode.
6. A network node (1, 2) according to any one of the claims 1-5, wherein the control unit (3, 4) is further arranged to add at least one Ethernet link to the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) used in the single logical link (7, 7A, 7B, 7C) if the actual data transfer load over the single logical link (7, 7A, 7B, 7C) is equal to or above a second predetermined threshold.
7. A network node (1, 2) according to claim 6, wherein the second predetermined threshold is at least partly based on the total data transfer capacity of the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) in the single logical link (7, 7A, 7B, 7C) before the addition of the at least one Ethernet link.
8. A network node (1, 2) according to claim 6 or 7, 7A, 7B, 7C, wherein the at least one added Ethernet link, prior to being added to the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) used in the single logical link (7, 7A, 7B, 7C), is in a standby idle mode.
9. A method for use in a network node (1, 2) for data communication over a plurality of Ethernet links (301, 302, 303, 304) comprising at least a first and a second Ethernet link, said method comprising the steps of: aggregating the at least first and second Ethernet link together in order to form a single logical link (7, 7A, 7B, 7C); and balancing the data transfer loads between the at least first and second Ethernet link comprised in the single logical link (7, 7A, 7B, 7C), characterized by
further comprising the steps of: monitoring the actual data transfer load over the single logical link (7, 7A, 7B, 7C); and controlling the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) used in the single logical link (7, 7A, 7B, 7C) at least partly based upon the actual data transfer load over the single logical link (7, 7A, 7B, 7C).
10. A method according to claim 9, further comprising the step of: removing at least one Ethernet link from the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) used in the single logical link (7, 7A, 7B, 7C) if the actual data transfer load over the single logical link (7, 7A, 7B, 7C) is equal to or below a first predetermined threshold; and/or
11. A method according to claim 10, further comprising the step of: re-distributing data traffic from the at least one Ethernet link to the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303,
304) remaining in the single logical link (7, 7A, 7B, 7C), for a predetermined time period prior to the removal of said at least one Ethernet link
12. A method according to any one of the claims 10-11, further comprising the step of: adding at least one Ethernet link to the actual number of Ethernet links of said plurality of Ethernet links (301, 302, 303, 304) used in the single logical link (7, 7A, 7B, 7C) if the actual data transfer load over the single logical link (7, 7A, 7B, 7C) is equal to or above a second predetermined threshold.
13. System for data communication over a plurality of Ethernet links (301, 302, 303, 304) comprising at least a first and a second Ethernet link, said system comprising at least one network node (1, 2) according any one of the claims 1-8.
14. A computer program product for use in a network node (1, 2), which comprises computer readable code means, which when run in a control unit (3, 4) in the network node (1, 2) causes the control unit (3, 4) to perform the steps of: - monitoring the actual data transfer load over a single logical link (7, 7 A, 7B,
7C); and controlling the actual number of Ethernet links of a plurality of Ethernet links (301, 302, 303, 304) used in the single logical link (7, 7A, 7B, 7C) at least partly based upon the actual data transfer load over the single logical link (7, 7A, 7B, 7C).
15. A computer program product according claim 14, comprising computer readable code means, which when run in the control unit (3, 4) in the network node (1, 2) causes the control unit (3, 4) to further perform the steps according to any one of the claims 10-12.
PCT/SE2009/050342 2009-03-31 2009-03-31 An energy efficient ethernet network node and a method for use in an ethernet network node WO2010114440A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/SE2009/050342 WO2010114440A1 (en) 2009-03-31 2009-03-31 An energy efficient ethernet network node and a method for use in an ethernet network node

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/SE2009/050342 WO2010114440A1 (en) 2009-03-31 2009-03-31 An energy efficient ethernet network node and a method for use in an ethernet network node

Publications (1)

Publication Number Publication Date
WO2010114440A1 true WO2010114440A1 (en) 2010-10-07

Family

ID=42828527

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/SE2009/050342 WO2010114440A1 (en) 2009-03-31 2009-03-31 An energy efficient ethernet network node and a method for use in an ethernet network node

Country Status (1)

Country Link
WO (1) WO2010114440A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2661048A1 (en) * 2012-05-04 2013-11-06 Huawei Technologies Co., Ltd Processing method and apparatus for member link in multilink group
CN112468393A (en) * 2020-11-04 2021-03-09 烽火通信科技股份有限公司 Method and device for selecting forwarding port based on link EEE and power consumption and electronic equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070201380A1 (en) * 2006-02-24 2007-08-30 Cisco Technology, Inc. Method and system for power-efficient adaptive link aggregation
US20080084828A1 (en) * 2006-10-09 2008-04-10 International Business Machines Corporation System and method for dynamically adding and deleting ethernet adapters
US20080304519A1 (en) * 2007-06-06 2008-12-11 Hewlett-Packard Development Company, L.P. Method for ethernet power savings on link aggregated groups

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070201380A1 (en) * 2006-02-24 2007-08-30 Cisco Technology, Inc. Method and system for power-efficient adaptive link aggregation
US20080084828A1 (en) * 2006-10-09 2008-04-10 International Business Machines Corporation System and method for dynamically adding and deleting ethernet adapters
US20080304519A1 (en) * 2007-06-06 2008-12-11 Hewlett-Packard Development Company, L.P. Method for ethernet power savings on link aggregated groups

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2661048A1 (en) * 2012-05-04 2013-11-06 Huawei Technologies Co., Ltd Processing method and apparatus for member link in multilink group
CN112468393A (en) * 2020-11-04 2021-03-09 烽火通信科技股份有限公司 Method and device for selecting forwarding port based on link EEE and power consumption and electronic equipment

Similar Documents

Publication Publication Date Title
US8595531B2 (en) Energy efficient ethernet network nodes and methods for use in ethernet network nodes
EP2184891B1 (en) Method and system for low latency state transitions for energy efficiency
EP2184890B1 (en) Method and system for control of energy efficiency and associated policies in a physical layer device
US20100115295A1 (en) Method And System For Energy Efficient Communication Among One Or More Interfaces In A Communication Path
CN101964713B (en) Network method and system
US20130003559A1 (en) Adaptive Power Savings for Aggregated Resources
US20070201380A1 (en) Method and system for power-efficient adaptive link aggregation
US8228796B2 (en) Ethernet switching apparatus, and method for reducing power consumption of the same
TWI513228B (en) System and method for energy efficient ethernet with asymmetric traffic profiles
US20050259685A1 (en) Dual speed interface between media access control unit and physical unit
KR102525573B1 (en) Power control method for power over data lines system
US9606615B2 (en) Apparatus, method and computer program means for data transport with reduced power consumption during link idle times
US9166868B2 (en) Distributed control plane for link aggregation
US20080084828A1 (en) System and method for dynamically adding and deleting ethernet adapters
EP3531627B1 (en) Auto-negotiation method and device for port
EP2717532B1 (en) High-speed link self-negotiation method and single board
WO2010114440A1 (en) An energy efficient ethernet network node and a method for use in an ethernet network node
CN111092739B (en) Dynamic adjusting method for power consumption of switch
CN101895729B (en) Streaming media server of embedded Linux shearing system
US9088490B2 (en) Network switching device, network switching system, and method of changing power state of network switching device
US20120188871A1 (en) Ethernet port speed control method and device
US8923290B1 (en) Self-forming lags
CN109639441A (en) A method of realizing that power consumption adjusts in real time based on business board
CN219812164U (en) Redundant switch
CN207801988U (en) A kind of Ethernet switch for supporting two kinds of redundancy application schemes

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09842777

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09842777

Country of ref document: EP

Kind code of ref document: A1