+

WO1998004066A1 - Distributed stack management - Google Patents

Distributed stack management Download PDF

Info

Publication number
WO1998004066A1
WO1998004066A1 PCT/GB1997/001999 GB9701999W WO9804066A1 WO 1998004066 A1 WO1998004066 A1 WO 1998004066A1 GB 9701999 W GB9701999 W GB 9701999W WO 9804066 A1 WO9804066 A1 WO 9804066A1
Authority
WO
WIPO (PCT)
Prior art keywords
communications
management
network
devices
stack
Prior art date
Application number
PCT/GB1997/001999
Other languages
French (fr)
Inventor
Peter Wilson
Peter Holiday
Richard Quine
Paul Garner
Graeme Mckerrell
Original Assignee
3Com Ireland
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 3Com Ireland filed Critical 3Com Ireland
Priority to EP97932947A priority Critical patent/EP0914734A1/en
Priority to GB9900425A priority patent/GB2330741B/en
Priority to JP50671898A priority patent/JP2001510644A/en
Publication of WO1998004066A1 publication Critical patent/WO1998004066A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks

Definitions

  • the present invention relates to the management of stacked communication devices, for instance in a local area network (LAN).
  • LAN local area network
  • computer networks such as LANs comprise a plurality of users which are enabled to communicate with each other via the network.
  • To enable such communication mere are provided a number of different types of interconnection device to which the users may be connected in order to form the network.
  • a communications device has a plurality of ports to each of which a user can be connected and the communications device provides links between the ports enabling the users to communicate with each other.
  • bridges and repeaters There are different types of known communication devices, for instances bridges and repeaters.
  • a bridge an assessment is made when an incoming communication is received on a port to determine the intended addressee of the communication, and the communication is sent out only the port or ports which are appropriate for the communication in question.
  • a repeater all communications received on any port are transmitted on all other ports so that all of the users on the network receive all of the communications issued by the other users and each user decides for itself whether to receive and/or to respond to the communication.
  • communications devices used in local area networks, for instance switches and routers, which act to connect users on the network in different ways and other devices connected in the network such as network analysers.
  • the details of all ese different devices are not important
  • management functions in the communications devices.
  • For the purposes of managing the network it is known to collect statistics relating to the communications which pass through the communications devices and mis is typically done by local collection of mese statistics and periodic sending of the collected statistics to a management device on interrogation. Also, it is possible that the communications devices may be remotely configured by the management device.
  • each communications device typically has a plurality of ports to which users may be connected and therefore a single such communications device can support a network up to that number of users. If it is desired to extend the capacity of the network beyond this number of users it is known to provide a stack of two or more communications devices each supporting a number of users of the network.
  • the plurality of communications devices are connected together so that users connected to one communications device can communicate with users connected to other devices. For the purposes of management in such a configuration, it is useful to be able to consider the stack of devices as a single entity for management purposes. This is advantageous in that the management device can simply send a single interrogation for statistics to the stack and receive back statistics relating to the whole stack.
  • a stack may be formed comprising a single master device and up to a predetermined number of slave devices.
  • the master device has a processor for communication with the management device and there are provided connections between the devices forming the stack to enable the processor within the master device to collect and store statistics relating to the operation of all the devices in the stack in order to be able to comply with interrogations from the management device.
  • This approach is disadvantageous for a number of reasons.
  • the communications which are required to occur between the master device and the slave devices require a substantial number of lines in the connection cable so that sufficient data from the slave devices can reach the master device for assessment by the processor.
  • the processor within the master device must be sufficiently large to be able to cope with the amount of data which may potentially be produced by the maximum number of slave units which may be connected to it. As the majority of networks would not be configured using the maximum number of slave devices with each master device this results in the provision of an over sized processor, at unnecessary cost.
  • a master device must either include enough processing power to cope with the maximum number of connections of slave units of different types, which is expensive, or it must be the case that only communication devices of the same type may be stacked together.
  • One development of the above arrangement has been to put a small micro-controller within each box in the stack, that micro-controller having basic management functions. There would still be a master device which would function to collect all the information for the stack overall for communication with the management device and for full management of the stack it is still the case that a Ml processor must be provided in the master device.
  • This further arrangement still suffers from many of the disadvantages mentioned above, in particular the fact that there is only one point of contact in the stack for the management device.
  • the present invention provides a method for the distributed management of a stack of communications devices of the general type described above.
  • each communications device is provided with sufficient processing ability to manage and collect data relating to communications occurring on its ports.
  • Each communications device is also provided with means to commumcate its collected data in a generic form, and preferably more than one includes means for receiving management communications and distributing these to all the devices in the stack.
  • the devices within the stack are interconnected by a simple communications link.
  • each communications device in the stack is provided with sufficient processing ability to be able to deal with its own ports. This means that when a network is set up, there is only provided the necessary processing power for managing the stack, rather than redundant processing power being provided to account for future expansion. Also, as each of the devices in the stack is identical from a management point of view the assembly of a stack by a user is considerably simplified as any number of devices may simply be attached together, without any consideration being given to the proper provision of a master unit with only a limited number of slave units. Also in the preferred arrangement of the present invention the management device may communicate with any one of a number of communications devices in the stack for interrogating the stack as a whole. This means that there is resilience in the management device connection and management does not become impossible if one device in the stack malfunctions.
  • the management of a stack of communications devices therefore provides advantages for the management of the network as a whole, as well as providing advantages in terms of cost and ease of use for a network user.
  • FIG. 1 illustrates in schematic form a plurality of communications devices 10 interconnected in a stack via the communications link 20.
  • Each communications device 10 comprises a plurality of ports 102 to each of which may be connected a user of the network.
  • Each communications device 10 may be any one of the known types of communications devices, for instance repeater, bridge, switch, router or any other type of device, eg dedicated network analyser, and provide interconnection between its ports 102 in a known fashion.
  • MAC media access control
  • Each device 10 also comprises a management device 120.
  • Management 120 comprises a processor 122, storage means 124 and communications means 126.
  • the processor means 122 is specific to the type of communications device 10 in which it is placed. That is, it provides sufficient processing power and function for the collection of the statistics and the management of the particular type of communications device (repeater, bridge, switch) to which it is attached.
  • the statistics collected by processor 122 are stored in storage means 124.
  • Each communications means 126 is capable of receiving communications from a remote management device via one of ports 102 and also of communicating with other communications device 126 and other devices 10 via communications link 20.
  • communications means 126 distributes the communication to the other devices 10 via communications link 20 and also responds to it itself.
  • management device 120 On receiving a management request from another box 10 via communications link 20 management device 120 considers the request itself and responds via communications link 20.
  • the communication of different types of communications devices 10 is facilitated because each unit 10 comprises sufficient processing power for the particular type of device which it is and also communications between devices 10 via communications link 20 are in a generic form so that it is not necessary for, for instance, the processor a repeater to be able to handle data specific to a bridge. It is simply the case that each communications means 126 must be sufficient to collate all the responses to a management request provided by all the boxes 10 in order to reply to the management unit.
  • each different type of communications device collects some different types of data.
  • the previous master/slave arrangement required the processor of the master unit to be able to cope with requests for all types of data for all types of devices.
  • any management request is simply copied to all the units, each of which responds to the appropriate parts of the request.
  • the communication means 26 in each device simply has to be able to handle all the replies, not understand them, and forward them to the management device.
  • each device 10 need only include sufficient processing power to be able to deal with its own ports and sufficient non-volatile memory for its own statistics.
  • each box should have the ability to provide generic coordination in order to send a full response from the whole stack back in response to a management request.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Small-Scale Networks (AREA)
  • Computer And Data Communications (AREA)

Abstract

A communication device for use in a network comprising a plurality of device units each having a plurality of ports and a communications link connecting the units together and for connection to the network, wherein each unit includes a processing means for managing and collecting data relating to communications on its ports only, means for generating messages in generic form relating to the data collected and means for applying the messages to the communication link.

Description

Distributed Stack Management
The present invention relates to the management of stacked communication devices, for instance in a local area network (LAN). As is well known computer networks such as LANs comprise a plurality of users which are enabled to communicate with each other via the network. To enable such communication mere are provided a number of different types of interconnection device to which the users may be connected in order to form the network. Typically a communications device has a plurality of ports to each of which a user can be connected and the communications device provides links between the ports enabling the users to communicate with each other.
There are different types of known communication devices, for instances bridges and repeaters. In a bridge an assessment is made when an incoming communication is received on a port to determine the intended addressee of the communication, and the communication is sent out only the port or ports which are appropriate for the communication in question. In a repeater, all communications received on any port are transmitted on all other ports so that all of the users on the network receive all of the communications issued by the other users and each user decides for itself whether to receive and/or to respond to the communication. There are other types of communications devices used in local area networks, for instance switches and routers, which act to connect users on the network in different ways and other devices connected in the network such as network analysers. However, for the purposes of understanding the present invention the details of all ese different devices are not important In addition to the basic functions of the network described above, there are typically additionally provided management functions in the communications devices. For the purposes of managing the network it is known to collect statistics relating to the communications which pass through the communications devices and mis is typically done by local collection of mese statistics and periodic sending of the collected statistics to a management device on interrogation. Also, it is possible that the communications devices may be remotely configured by the management device.
As mentioned above, each communications device typically has a plurality of ports to which users may be connected and therefore a single such communications device can support a network up to that number of users. If it is desired to extend the capacity of the network beyond this number of users it is known to provide a stack of two or more communications devices each supporting a number of users of the network. The plurality of communications devices are connected together so that users connected to one communications device can communicate with users connected to other devices. For the purposes of management in such a configuration, it is useful to be able to consider the stack of devices as a single entity for management purposes. This is advantageous in that the management device can simply send a single interrogation for statistics to the stack and receive back statistics relating to the whole stack.
One previous approach to providing this function has been to provide two basic types of communications devices which may be formed in a stack, the two types being designated "master" and "slave". In this arrangement a stack may be formed comprising a single master device and up to a predetermined number of slave devices. The master device has a processor for communication with the management device and there are provided connections between the devices forming the stack to enable the processor within the master device to collect and store statistics relating to the operation of all the devices in the stack in order to be able to comply with interrogations from the management device. This approach is disadvantageous for a number of reasons.
Firstly, the communications which are required to occur between the master device and the slave devices require a substantial number of lines in the connection cable so that sufficient data from the slave devices can reach the master device for assessment by the processor. Secondly, the processor within the master device must be sufficiently large to be able to cope with the amount of data which may potentially be produced by the maximum number of slave units which may be connected to it. As the majority of networks would not be configured using the maximum number of slave devices with each master device this results in the provision of an over sized processor, at unnecessary cost. Thirdly, in such an arrangement there is only one point of contact for the management device to commumcate with the stack, and thus if the master device malfunctions, there is no way for the management device to communicate. Fourthly, as different types of communication devices (eg bridges, repeaters, switches) require different amounts of processing and different types of data collection in this arrangement a master device must either include enough processing power to cope with the maximum number of connections of slave units of different types, which is expensive, or it must be the case that only communication devices of the same type may be stacked together. One development of the above arrangement has been to put a small micro-controller within each box in the stack, that micro-controller having basic management functions. There would still be a master device which would function to collect all the information for the stack overall for communication with the management device and for full management of the stack it is still the case that a Ml processor must be provided in the master device. This further arrangement still suffers from many of the disadvantages mentioned above, in particular the fact that there is only one point of contact in the stack for the management device.
Further, in either of the above mentioned systems, it is possible mat communications which take place between ports on one slave device may not be properly accounted for in the collection of the statistics by the master device.
The present invention provides a method for the distributed management of a stack of communications devices of the general type described above. According to this method each communications device is provided with sufficient processing ability to manage and collect data relating to communications occurring on its ports. Each communications device is also provided with means to commumcate its collected data in a generic form, and preferably more than one includes means for receiving management communications and distributing these to all the devices in the stack. Finally, according to the method of this invention the devices within the stack are interconnected by a simple communications link.
In the present invention therefore each communications device in the stack is provided with sufficient processing ability to be able to deal with its own ports. This means that when a network is set up, there is only provided the necessary processing power for managing the stack, rather than redundant processing power being provided to account for future expansion. Also, as each of the devices in the stack is identical from a management point of view the assembly of a stack by a user is considerably simplified as any number of devices may simply be attached together, without any consideration being given to the proper provision of a master unit with only a limited number of slave units. Also in the preferred arrangement of the present invention the management device may communicate with any one of a number of communications devices in the stack for interrogating the stack as a whole. This means that there is resilience in the management device connection and management does not become impossible if one device in the stack malfunctions.
When a particular device receives a management commumcation it distributes it to the other devices in the stack and each device provides its own response in the generic form. The device which receives the interrogation request then simply collates the answers and forwards them to the management device. The management of a stack of communications devices according to the present invention therefore provides advantages for the management of the network as a whole, as well as providing advantages in terms of cost and ease of use for a network user.
This invention will be better understood from the following description of a preferred embodiment given by way of example and with reference to the accompanying Figure 1.
Figure 1 illustrates in schematic form a plurality of communications devices 10 interconnected in a stack via the communications link 20. Each communications device 10 comprises a plurality of ports 102 to each of which may be connected a user of the network. Each communications device 10 may be any one of the known types of communications devices, for instance repeater, bridge, switch, router or any other type of device, eg dedicated network analyser, and provide interconnection between its ports 102 in a known fashion. In addition to those provided within each communications device for the performance of its communications function, there is provided a media access control (MAC) unit in each device for connection to the communications link 20. This means that the communications link can be implemented simply as a network segment for example as a 10Mbps Ethernet bus. This simplifies the implementation of the communications link 20.
Each device 10 also comprises a management device 120. Management 120 comprises a processor 122, storage means 124 and communications means 126. The processor means 122 is specific to the type of communications device 10 in which it is placed. That is, it provides sufficient processing power and function for the collection of the statistics and the management of the particular type of communications device (repeater, bridge, switch) to which it is attached. The statistics collected by processor 122 are stored in storage means 124.
Each communications means 126 is capable of receiving communications from a remote management device via one of ports 102 and also of communicating with other communications device 126 and other devices 10 via communications link 20. In broad terms, on receiving a management communication via a port 102, communications means 126 distributes the communication to the other devices 10 via communications link 20 and also responds to it itself. On receiving a management request from another box 10 via communications link 20 management device 120 considers the request itself and responds via communications link 20. In the present invention the communication of different types of communications devices 10 is facilitated because each unit 10 comprises sufficient processing power for the particular type of device which it is and also communications between devices 10 via communications link 20 are in a generic form so that it is not necessary for, for instance, the processor a repeater to be able to handle data specific to a bridge. It is simply the case that each communications means 126 must be sufficient to collate all the responses to a management request provided by all the boxes 10 in order to reply to the management unit.
In the collection of statistics it is the case that each different type of communications device collects some different types of data. In a stack which includes a number of different types of device, the previous master/slave arrangement required the processor of the master unit to be able to cope with requests for all types of data for all types of devices. In the present arrangement however any management request is simply copied to all the units, each of which responds to the appropriate parts of the request. The communication means 26 in each device simply has to be able to handle all the replies, not understand them, and forward them to the management device.
This invention is advantageous in terms of the cost of each device 10 in that each such device need only include sufficient processing power to be able to deal with its own ports and sufficient non-volatile memory for its own statistics. The only additional requirement is that each box should have the ability to provide generic coordination in order to send a full response from the whole stack back in response to a management request.

Claims

CLAIMS:
1. A commumcation device for use in a network comprising a plurality of device units each having a plurality of ports and a communications link connecting the units together and for connection to the network, wherein each unit includes a processing means for managing and collecting data relating to communications on its ports only, means for generating messages in generic form relating to the data collected and means for applying the messages to the communication link.
2. A communication device according to claim 1, wherein more than one of the units is provided with means for receiving management communications from the network and distributing the received communications to all units linked by the communications link.
3. A commumcation device according to claim 2, wherein one of said units provided with means for receiving management communications is arranged to collate and forward answers from the other units in response to an interrogation request.
PCT/GB1997/001999 1996-07-23 1997-07-23 Distributed stack management WO1998004066A1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP97932947A EP0914734A1 (en) 1996-07-23 1997-07-23 Distributed stack management
GB9900425A GB2330741B (en) 1996-07-23 1997-07-23 Distributed stack management
JP50671898A JP2001510644A (en) 1996-07-23 1997-07-23 Distributed stack management

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
GB9615423.2 1996-07-23
GBGB9615423.2A GB9615423D0 (en) 1996-07-23 1996-07-23 Distributed stack management

Publications (1)

Publication Number Publication Date
WO1998004066A1 true WO1998004066A1 (en) 1998-01-29

Family

ID=10797346

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB1997/001999 WO1998004066A1 (en) 1996-07-23 1997-07-23 Distributed stack management

Country Status (4)

Country Link
EP (1) EP0914734A1 (en)
JP (1) JP2001510644A (en)
GB (1) GB9615423D0 (en)
WO (1) WO1998004066A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2333429A (en) * 1997-11-28 1999-07-21 3Com Technologies Ltd Stacked communication devices
US6801953B2 (en) 1997-11-28 2004-10-05 3Com Corporation Trunking in stacked communication devices

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5226120A (en) * 1990-05-21 1993-07-06 Synoptics Communications, Inc. Apparatus and method of monitoring the status of a local area network

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5226120A (en) * 1990-05-21 1993-07-06 Synoptics Communications, Inc. Apparatus and method of monitoring the status of a local area network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
VASSILA A ET AL: "INTRODUCING ACTIVE MANAGED OBJECTS FOR EFFECTIVE AND AUTONOMOUS DISTRIBUTED MANAGEMENT", BRINGING TELECOMMUNICATION SERVICES TO THE PEOPLE - ISS & N 1995, THIRD INTERNATIONAL CONFERENCE ON INTELLIGENCE IN BROADBAND SERVICE AND NETWORKS, HERAKLION, CRETE, OCT. 16 - 19, 1995. PROCEEDINGS, no. CONF. 3, 16 October 1995 (1995-10-16), CLARKE A;CAMPOLARGO M; KARATZAS N (EDS ), pages 415 - 429, XP000593492 *
YECHIAM YEMINI: "THE OSI NETWORK MANAGEMENT MODEL", IEEE COMMUNICATIONS MAGAZINE, vol. 31, no. 5, 1 May 1993 (1993-05-01), pages 20 - 29, XP000367615 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2333429A (en) * 1997-11-28 1999-07-21 3Com Technologies Ltd Stacked communication devices
US6425015B1 (en) 1997-11-28 2002-07-23 3 Com Technologies Stacked communication devices and method for port mirroring using modified protocol
GB2333429B (en) * 1997-11-28 2002-11-13 3Com Technologies Ltd Stacked communication devices
US6801953B2 (en) 1997-11-28 2004-10-05 3Com Corporation Trunking in stacked communication devices

Also Published As

Publication number Publication date
JP2001510644A (en) 2001-07-31
EP0914734A1 (en) 1999-05-12
GB9615423D0 (en) 1996-09-04

Similar Documents

Publication Publication Date Title
US5420857A (en) Connection establishment in a flat distributed packet switch architecture
US5295133A (en) System administration in a flat distributed packet switch architecture
US5991828A (en) System for automatically connecting portable device to network using network environment information including domain name of naming device and community name of network management protocol
AU675362B2 (en) Determination of network topology
US6292472B1 (en) Reduced polling in an SNMPv1-managed network
EP1146429A1 (en) Modular computer backplane switch with dynamic bus allocation
JP4202026B2 (en) Storage system and storage device
CA2092966A1 (en) Distribution of communication connections over multiple service access points in a communications network
EP0186320A1 (en) Local area communication network
US6311208B1 (en) Server address management system
US7558565B2 (en) Methods and apparatus for wireless network load generator clustering
EP0914734A1 (en) Distributed stack management
CN1316789C (en) Improvements in and related to telecommunications networks
KR100663579B1 (en) Asynchronous transmission method (ATM) performance data real-time collection device and method thereof in high-speed information communication network
EP1114546B1 (en) Transmission system adapted for ip data packets
US6836802B1 (en) Network management system uses proxy agent for communicating control information and event notifications between network management apparatus and agents not directly connected to a network
JPH0125464B2 (en)
JPH1155256A (en) Reconnection processing method in multipoint connection
JP4602423B2 (en) Storage device and storage system
JP2743890B2 (en) Network management method
KR100771915B1 (en) Digital Protection Relay and Communication Redundancy Method
JPH08154093A (en) Radio communication system
JP3836301B2 (en) Heterogeneous network connection device and heterogeneous network connection method
US6542505B1 (en) ATM switching apparatus with server function
JP2730094B2 (en) Mailbox control method for electronic mail system

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): GB JP US

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 1997932947

Country of ref document: EP

ENP Entry into the national phase

Ref country code: GB

Ref document number: 9900425

Kind code of ref document: A

Format of ref document f/p: F

ENP Entry into the national phase

Ref country code: JP

Ref document number: 1998 506718

Kind code of ref document: A

Format of ref document f/p: F

WWP Wipo information: published in national office

Ref document number: 1997932947

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 09214994

Country of ref document: US

WWW Wipo information: withdrawn in national office

Ref document number: 1997932947

Country of ref document: EP

点击 这是indexloc提供的php浏览器服务,不要输入任何密码和下载