[go: up one dir, main page]

WO2008148173A1 - Système et procédé servant à améliorer un débit dans un dispositif de réseau - Google Patents

Système et procédé servant à améliorer un débit dans un dispositif de réseau Download PDF

Info

Publication number
WO2008148173A1
WO2008148173A1 PCT/AU2008/000826 AU2008000826W WO2008148173A1 WO 2008148173 A1 WO2008148173 A1 WO 2008148173A1 AU 2008000826 W AU2008000826 W AU 2008000826W WO 2008148173 A1 WO2008148173 A1 WO 2008148173A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
accordance
acknowledgement
packet
preferred rate
Prior art date
Application number
PCT/AU2008/000826
Other languages
English (en)
Inventor
Bjorn Gustaf Landfeldt
Original Assignee
Smart Internet Technology Crc Pty Ltd
Armitage, Grenville
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
Priority claimed from AU2007903063A external-priority patent/AU2007903063A0/en
Application filed by Smart Internet Technology Crc Pty Ltd, Armitage, Grenville filed Critical Smart Internet Technology Crc Pty Ltd
Publication of WO2008148173A1 publication Critical patent/WO2008148173A1/fr

Links

Classifications

    • 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
    • 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
    • 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
    • H04L47/19—Flow control; Congestion control at layers above the network layer
    • H04L47/193—Flow control; Congestion control at layers above the network layer at the transport layer, e.g. TCP related
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2466—Traffic characterised by specific attributes, e.g. priority or QoS using signalling traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/25—Flow control; Congestion control with rate being modified by the source upon detecting a change of network conditions

Definitions

  • the present invention generally relates to a system and method for improving throughput in a network device .
  • Embodiments of the invention find particular, but not exclusive, use in asymmetric devices, such as an Asymmetric Digital Subscriber Line (ADSL) modem device.
  • ADSL Asymmetric Digital Subscriber Line
  • TCP Transmission Control Protocol
  • the term 'data' includes emails, data files (whether as email attachments or raw files) , web pages, objects inside web pages (such as images) , non-interactive video or audio transmissions, etc. That is, the term ⁇ data' can be construed to include any type of electronic information that is capable of being sent over an electronic network (including a wireless network) .
  • the TCP standard automatically adjusts the data transmission rate to best suit the apparent network capacity between source and destination. TCP adjusts the transmission rate by attempting to measure (or estimate) apparent network capacity. This is achieved by sending data packets to a destination and monitoring the stream of returned Acknowledgment (ACK) packets from the destination. TCP adjusts the transmission of data based on the rate at which matching ACK packets return from the destination.
  • TCP detects packet loss events by tracking the non-receipt of matching ACK packets. TCP responds to loss events by repeating transmission of lost data packets, and temporarily slowing down the transmission of subsequent data packets.
  • Network behaviour that interferes with or constrains the free flow of ACK packets from destination to source is known to cause suboptimal data transfer performance from source to destination.
  • the present invention provides a system for a method for managing data packets to be sent through an asymmetric network device, comprising the steps of receiving data from at least one computing device, determining whether the data is acknowledgement data, and if so, prioritising the upload of the acknowledgement data to the asymmetric network device .
  • the method may include the initial step of determining a preferred rate of acknowledgement data to be uploaded to the asymmetric network device, and prioritising the upload of the acknowledgement data until the preferred rate is reached.
  • the preferred rate may be predetermined (by a user, for example) , or it may be calculated dynamically. Where the preferred rate is calculated dynamically, the dynamic calculation may be a function of the instantaneous download rate, the data packet size and/or the acknowledgement data packet size.
  • U m i n is the preferred rate
  • B 3 is the average size of an acknowledgement packet
  • B d is the average size of a data packet
  • F is a predetermined fraction set by a user
  • D is the maximum download capacity through the network device.
  • B a may be calculated utilising the formula
  • B avg is the network defined average size of an acknowledgment packet
  • X is the number of data packets downloaded for every acknowledgment packet uploaded. All received data may be placed into a queuing structure, such that acknowledgement data is placed in one queue, and the acknowledgement data queue is prioritised.
  • the step of determining whether data is acknowledgement data may include the step of, inspecting the header within the data packet to determine whether the data is acknowledgement data and/or inspecting the size of the data packet to determine whether the data is acknowledgement data.
  • the present invention provides a method for a system for managing a device connected to an asymmetric network device, comprising a receiving module arranged to receive data from at least one computing device, and a prioritisation module arranged determine whether the data is acknowledgement data, and if so, prioritise the upload of the data to the asymmetric network device .
  • the present invention provides a computing program including at least one instruction which, when loaded on a programmable device, causes the programmable device to perform a method in accordance with the second aspect of the invention.
  • the present invention provides a computer readable medium incorporating a computing program in accordance with a third aspect of the invention.
  • Figure 1 is a diagram illustrating a computing system suitable for implementing a methodology or system in accordance with an embodiment of the present invention.
  • Figure 2 is a flowchart illustrating a methodology in accordance with an embodiment of the present invention.
  • FIG. 1 there is shown a schematic diagram of a device 100 suitable for use with an embodiment of the present invention.
  • the device 100 may be used to execute applications and/or system services such as a throughput improvement algorithm or methodology in accordance with an embodiment of the present invention.
  • the device 100 may comprise suitable components necessary to receive, store and execute appropriate computing instructions.
  • the components may include a processor 102, read only memory (ROM) 104, random access memory (RAM) 106, and input/output devices such as disk drives 108, input device 110 (such as an Ethernet port, a USB port, etc.), display 112 (such as a liquid crystal display, a light emitting diode display, or any other suitable display) and communications link 114.
  • the device includes instructions that may be stored in ROM 104, RAM 106, or disk drives 108 and may be executed by the processor 102.
  • There may be provided a plurality of communications links 114 which may variously connect to one or more computing devices, such as servers, personal computers, terminals, wireless or handheld computing devices. At least one of the plurality of communications links may be connected to an external computing network through a telephone line or other type of communications link.
  • the device may include persistent storage devices such as disk drives 108 which may encompass solid state drives, hard disk drives, optical drives or magnetic tape drives.
  • the device 100 may- use a single disk drive or multiple disk drives.
  • the device 100 may use a suitable operating system.
  • the device 100 may be a router or gateway device arranged to receive data packets from an asymmetric source such as an asymmetric digital subscription line modem which is arranged to receive data from an external network.
  • an asymmetric source such as an asymmetric digital subscription line modem which is arranged to receive data from an external network.
  • the device 100 includes a software application 116 (or instructions) which cause the device to implement a method or computer program in accordance with an embodiment of the invention.
  • an asymmetric device is one which exhibits more capacity (greater bandwidth) downstream (towards the local client) and less capacity (lower bandwidth) upstream (out towards the service provider's network, or the Internet) .
  • TCP standard's ability to send data packets downstream over an asymmetric link is limited not only by the physical capacity of the link, but also by the rate at which ACK packets can be sent upstream over that same link.
  • TCP Data packets are typically far larger than ACK packets (often 1500 bytes for full Data packets compared to 40 bytes for typical ACK packets) .
  • the downlink speed is denoted as D (bytes/sec)
  • the uplink speed is denoted as U (bytes/sec) .
  • D bytes/sec
  • U bytes/sec
  • the minimum value of the uplink speed U m i n required to fully utilise downlink capacity D can be calculated to be approximately:
  • the TCP protocol is unable to fully utilise all of the downlink capacity D for data transmission, regardless of how large D becomes, due to the inability of any uplink sites being able to receive the requisite number of ACK packets.
  • the downstream transmission rate will be limited to approximately:
  • U ack may be limited to less than U where the upstream link is being shared by multiple flows of packets, where not all of the ACK packets relate to the TCP session of interest. Therefore, the downstream transmission rate, in the presence of competing traffic in the upstream direction, will be limited to (U ac k/40) *1500 bytes/second.
  • the need to send data through the uplink i.e. data other than ACK packets
  • the upstream link may be shared amongst various applications running on one or more computing devices that are all connected to the LAN.
  • Some applications may require substantial fractions of the capacity in the upstream link.
  • the amount of uplink bandwidth available for the sending of ACK packets decreases.
  • the interface between the LAN and the broadband access service is a router (or gateway) which accepts packets to be transmitted on a first-come-first- served basis, U ack can be forced to an exceedingly low fraction of U, as the traffic is competing for the upstream link on a first-come-first-served basis.
  • a router/gateway (such as the device described with reference to Figure 1) can be programmed to implement a manually configurable scheme for providing minimum bandwidth guarantees to particular traffic in the upstream direction.
  • the embodiment described herein provides a method and system for configuring the device 100 to provide (i.e. reserve) a predetermined bandwidth of upstream U for the transmission of TCP ACK packets. That is, the embodiment contemplates determining a value for U aCk and embodying the determined value in the device in a manner which ensures that a certain amount of ACK packets are correctly delivered upstream.
  • the predetermined value of U aCk remains fixed. That is, U aCk is set to a value that remains constant irrespective of other traffic demands. This value may be set permanently (for example, by an Internet Service Provider or Network Administrator) or may be varied by an end user, through a software application.
  • the device 100 determines or is given a predetermined value for U aCk - Then, as traffic is received by the device 205, ACK packets are identified 210, and prioritised 215 so that the packets are sent upstream with minimal delay.
  • U aCk is optimal, which in the example described herein, is denoted as U m i n (i.e. the optimal minimum bandwidth reserved for ACK packets)
  • the value is recalculated on a periodic basis, to ensure that the download capacity D is optimised at all times.
  • the value of D can fluctuate from time to time as the access service adjusts physical layer parameters to suit line conditions or where contractual parameters set by the customer's internet service provider are changed (for example, a user upgrades from a 256/64k plan to a 1024/128k plan) . Therefore, in order to calculate an optimal value of U m i n , the actual, current value of D must be known.
  • the value of D is obtained by probing the downstream physical layer circuitry built into the device 100. Many routers/gateways such as device 100 provide such information through a standard software tool or library function.
  • the value of U m i n is calculated to allow TCP downstream data traffic to utilise a predetermined optimal fraction (F) of the downstream capacity D.
  • U tn in (B a /B d )*F*D
  • B a is the average size of TCP ACK packets
  • B d is the average size of TCP Data packets in bytes.
  • the term 'optimal' is utilised to denote a condition that is most desirable or advantageous to a user or network.
  • the optimal value may be a value of 1 (i.e. 100% usage of the download capacity) .
  • the optimal value may be a fraction less than 1 (i.e. under 100% usage of download capacity) .
  • an optimal value may not necessarily be the maximum value theoretically available, but may be the maximum value available in light of other constraints, such as the need to reserve uplink bandwidth for other applications .
  • the embodiment described herein also contemplates periodically recalculating B a and Ba as the device detects changes in the TCP MSS (maximum segment size) of the downstream TCP traffic. This ensures that U m i n is optimised.
  • B a may be adjusted to take into account particular network implementations. For example, some TCP implementations may send one ACK packet for every two TCP data packets (rather than one ACK per TCP data packet) . The amount of upstream bandwidth required to optimally utilise the downstream bandwidth is therefore halved, and correspondingly, the upload bandwidth required to ensure optimal download is also changed.
  • the variation in the number of ACK packets uploaded can be accounted for dynamically through the introduction of an optional scaling factor to change the value of B a (the size of an ACK packet) .
  • B a is taken to be 40 bytes if one 40 byte ACK packet is uploaded up for each TCP data packet downloaded. However, if one 40 byte ACK packet is uploaded for every two TCP data packets downloaded, then B a can be scaled to 20 bytes to take into account the reduced bandwidth requirements for ACK packets .
  • the value of B a can be calculated utilising a mathematical relationship such as :
  • X is the number of TCP data packets downloaded for each ACK packet that is uploaded
  • B avg is 40 bytes, the size of the average ACK packet.
  • the calculation can also be extended to cover cases where D and U m i n do not necessarily vary in a linear manner (e.g. where an Asynchronous Transfer Mode layer sits between the packet layer and the physical link layer) .
  • One method for determining the types of packets received by the device 100 is to inspect upstream packets in detail to determine if they are TCP ACK packets. This may be done by inspecting the header of each TCP packet which arrives at device 100.
  • all TCP packets of a given size may be forwarded preferentially through the reserved bandwidth.
  • the router may exclude from U m j .n any ACK packets that piggyback on larger upstream TCP Data packets going in the reverse direction.
  • ACK packet size is fixed
  • ACK packets that piggyback on larger upstream TCP Data packets may be ignored. This may be done consciously for the sake of simplicity.
  • the device may also identify upstream Data packets that carry an ACK signal for associated downstream packets as packets which should be forwarded preferentially.
  • the ACK packets are forwarded through the uplink in a preferential manner. In other words, the ACK packets receive priority.
  • the prioritisation of ACK packets may be implemented by utilising any suitable form of queuing system which is arranged to provide the ACK packets with some 'preference' over other packets.
  • One methodology is differential queuing and scheduling, which prioritises any ACK packets sent on the upstream link, to a guaranteed minimum of U m i n bytes per second.
  • packets may be placed in one of two main queues- ACK or non-ACK.
  • the queuing algorithm gives the ACK queue absolute preferential treatment over the non-ACK queue, but only to the guaranteed U m i n value. It will be understood that other queuing algorithms which provide ACK packets with priority may also be utilised. Such variations are within the purview of a person skilled in the art.
  • other upstream traffic is capable of utilising all of the available U during periods where there is no TCP ACK traffic.
  • the embodiment described herein provides a number of advantages. Firstly, the embodiment may be implemented in either software or firmware, as desired, and is retro- fittable to many network devices such as routers and gateways .
  • the embodiment while being capable of being implemented as part of a network standard or protocol, does not require changes to existing standards or protocols in order to operate. Therefore, the embodiment can work harmoniously with existing standards, while being capable of being integrated into future standards .
  • the methodology and system optimises download bandwidth on asymmetric devices, thereby enhancing user experience and user satisfaction. This is particularly important in networks and jurisdictions where a large percentage of users utilise asymmetric communication devices.
  • asymmetric communication devices One example is Australia, where a large number of Internet users are connected through ADSL modems or satellite links.
  • the incremental impact on TCP throughput from the RTT (round trip time) increase perceived by ACKs that are forced to be queued on a congested uplink is improved by ensuring that there is less or no congestion of ACK packets.
  • the decreased need to resend data reduces congestion throughout the network as a whole, thereby potentially providing more available bandwidth for all users on the network.
  • the embodiments described with reference to the Figures can be implemented via an application programming interface (API) or as a series of libraries, for use by a developer, and can be included within another software application, such as a terminal or personal computer operating system or a portable computing device operating system.
  • API application programming interface
  • program modules include routines, programs, objects, components, and data files that perform or assist in the performance of particular functions, it will be understood that the functionality of the software application may be distributed across a number of routines, objects and components yet achieve the same functionality as the embodiment and the broader invention claimed herein. Such variations and modifications are within the purview of those skilled in the art.

Landscapes

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

Abstract

La présente invention concerne un système et un procédé servant à gérer des paquets de données devant être envoyés par le biais d'un dispositif de réseau asymétrique. Le procédé comprend les étapes consistant à recevoir des données d'au moins un dispositif informatique, et à déterminer si les données sont des données d'accusé de réception. Si c'est le cas, le téléchargement des données d'accusé de réception est mis en priorité sur le dispositif de réseau asymétrique.
PCT/AU2008/000826 2007-06-07 2008-06-10 Système et procédé servant à améliorer un débit dans un dispositif de réseau WO2008148173A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
AU2007903063A AU2007903063A0 (en) 2007-06-07 A system and method for improving throughput in a network device
AU2007903063 2007-06-07

Publications (1)

Publication Number Publication Date
WO2008148173A1 true WO2008148173A1 (fr) 2008-12-11

Family

ID=40093092

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/AU2008/000826 WO2008148173A1 (fr) 2007-06-07 2008-06-10 Système et procédé servant à améliorer un débit dans un dispositif de réseau

Country Status (1)

Country Link
WO (1) WO2008148173A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8194543B2 (en) 2008-12-18 2012-06-05 Intel Mobile Communications GmbH Methods of data traffic shaping, apparatus and wireless device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5495481A (en) * 1994-09-30 1996-02-27 Apple Computer, Inc. Method and apparatus for accelerating arbitration in a serial bus by detection of acknowledge packets
US6424626B1 (en) * 1999-10-29 2002-07-23 Hubbell Incorporated Method and system for discarding and regenerating acknowledgment packets in ADSL communications
EP1303072A2 (fr) * 2001-10-15 2003-04-16 Texas Instruments Incorporated Téléchargement amélioré pour ADSL avec file d'attente de priorités de transmission
US20070002863A1 (en) * 2005-06-14 2007-01-04 Microsoft Corporation Multi-stream acknowledgement scheduling

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5495481A (en) * 1994-09-30 1996-02-27 Apple Computer, Inc. Method and apparatus for accelerating arbitration in a serial bus by detection of acknowledge packets
US6424626B1 (en) * 1999-10-29 2002-07-23 Hubbell Incorporated Method and system for discarding and regenerating acknowledgment packets in ADSL communications
EP1303072A2 (fr) * 2001-10-15 2003-04-16 Texas Instruments Incorporated Téléchargement amélioré pour ADSL avec file d'attente de priorités de transmission
US20070002863A1 (en) * 2005-06-14 2007-01-04 Microsoft Corporation Multi-stream acknowledgement scheduling

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
BALAKRISHNAN ET AL.: "The effects of asymmetry on TCP performance", MOBILE NETWORKS AND APPLICATIONS, September 1999 (1999-09-01), pages 219 - 240, XP000875880 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8194543B2 (en) 2008-12-18 2012-06-05 Intel Mobile Communications GmbH Methods of data traffic shaping, apparatus and wireless device
US8755278B2 (en) 2008-12-18 2014-06-17 Intel Mobile Communications GmbH Methods of data traffic shaping, apparatus and wireless device
DE102009044647B4 (de) * 2008-12-18 2014-10-23 Intel Mobile Communications GmbH Verfahren zur Datenverkehrsflusssteuerung, Vorrichtung und Drahtlos-Gerät

Similar Documents

Publication Publication Date Title
EP2992652B1 (fr) Gestion d'attribution de bande passante parmi des flux par le biais d'assignation de priorité de suppression
US20220417151A1 (en) Packet transmission system and method
EP1535130B1 (fr) Gestion de la qualite de service dans des passerelles de reseau
EP2997707B1 (fr) Amélioration des changements de canaux rapides et autres changements de position de lecture dans la lecture en transit adaptative
EP3044918B1 (fr) Limitation de vitesse adaptative basée sur un réseau
US8259566B2 (en) Adaptive quality of service policy for dynamic networks
EP2438716B1 (fr) Mesure de trafic en fonction de la congestion
JP7638537B2 (ja) データパケット通信を管理するためのシステムおよび方法
WO2005107192A1 (fr) Systeme et procede d'amelioration de la qualite de service d'un reseau
EP2594043A1 (fr) Système, procédé et programme informatique pour la distribution intelligente de paquets
US20070248101A1 (en) Efficient policer based weighted fair bandwidth method and system
KR101837637B1 (ko) 클라이언트 측 ack 조정 기반 적응 스트리밍 방법 및 장치
EP3442182B1 (fr) Chute ou marquage de joint de paquets de données de contrôle de congestion du trafic
JP2002532959A (ja) ワイヤレスローカルループシステムおよびそれに役立つ方法
WO2008148173A1 (fr) Système et procédé servant à améliorer un débit dans un dispositif de réseau
EP2388978B1 (fr) Procédés et dispositifs pour déterminer la charge de liaisons dans un réseau
Yoon et al. Weighted bandwidth sharing scheme to guarantee the video quality in home networks
Fetoh et al. A Framework for Video Transmission Scheduling Using Concurrent Multipath Transfer
Lee et al. Congestion control scheme for UHD video streaming over the Internet
Takeo et al. Application-level QoS of Web access and streaming services with AF service on DiffServ
Jain et al. Review of Quality of Servie (QoS) Issues in Mobile Ad hoc Networks.

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: 08756909

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATE 17/03/2010)

122 Ep: pct application non-entry in european phase

Ref document number: 08756909

Country of ref document: EP

Kind code of ref document: A1