GB2294340A - Communications system design. - Google Patents

Communications system design. Download PDF

Info

Publication number
GB2294340A
GB2294340A GB9421227A GB9421227A GB2294340A GB 2294340 A GB2294340 A GB 2294340A GB 9421227 A GB9421227 A GB 9421227A GB 9421227 A GB9421227 A GB 9421227A GB 2294340 A GB2294340 A GB 2294340A
Authority
GB
United Kingdom
Prior art keywords
components
component
configuration
products
services
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
GB9421227A
Other versions
GB9421227D0 (en
Inventor
Andrew Tunnicliffe
Clive C Hayball
Kevin Lewis
Robin Andrew Lewis
Mark Brendan Ward
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northern Telecom Ltd
Original Assignee
Northern Telecom Ltd
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 Northern Telecom Ltd filed Critical Northern Telecom Ltd
Priority to GB9421227A priority Critical patent/GB2294340A/en
Publication of GB9421227D0 publication Critical patent/GB9421227D0/en
Publication of GB2294340A publication Critical patent/GB2294340A/en
Application status is Withdrawn legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/50Computer-aided design

Abstract

A method for designing a telecommunications system which defines components in the form of telecommunication products and telecommunication services. Such components are defined in terms of the resources consumed by the products and services and the supplies provided by the products and services. Products and components are defined with no implicit or explicit references to other components so that they only provide and/or make use of resources. <IMAGE>

Description

COMMUNICATIONS SYSTEM DESIGN METHOD AND DESIGN TOOL THEREFOR FIELD OF THE INVENTION This invention relates to a communications system design method for designing a communications system operable for specific user requirements. Such a system will be configured with known telecommunications equipment having particular characteristics and known telecommunications services also having particular characteristics.

BACKGROUND OF THE INVENTION The technology implemented in telecommunications is fundamentally knowledge based. A knowledge based system is one in which a clear separation exists between the description of the problem domain and the reasoning process over that knowledge. The knowledge base is an intuitive representation of the problem domain that facilitates ease of acquisition and maintenance. In the design of a telecommunications system e.g. a local area network, a user will require a variety of equipment, such as telephones, a computer network complete with hard disc, printers, modems, video links and the like. These items will be connected in a particular fashion such that optimal use is made of the resources, communication channels between rooms in a building , areas within a room and between associated buildings and sites.Broadly speaking design or configuration can be considered as a creation of a typical system according to a requirement specification.

Configuration can thus be considered as an architecture that defines certain ways of connecting a pre-defined set of components to meet the demands imposed by a user for a specific situation whereby a set of components that satisfies all relevant requirements is selected or inconsistencies in the requirement are detected. This is enabled by selecting, parameterizing and positioning instances of suitable existing component types from a given catalogue.

In performing the configuration, access will be required to product knowledge which describes the performance of each product1 the engineering rules associated with each product, and evluation criteria which govern the ways of using product knowledge to perform configuration. The configuration task is one that takes the human expert considerable time to perform. Many approaches and tools are currently available. It is common for new configuration tools to be developed for each and every new product or technology range. Developments of the product range, however, often causes such tools to get left behind and either require expensive modification or simply become unusable.

Traditional configuration systems have used a knowledge representation scheme where problem solving knowledge and component knowledge have been intermingled, leading to difficulties in knowledge base maintenance.

EP-A3-0540487 (Televerket) relates to a system design method for use in designing a product, system or service and is able to support services.

This document describes the construction of an IT system and employs a user model of the product. The design procedure is a process of topdown interface design which entails decomposing service states and tasks in the design of system objects and functions. This process is structured with the aid of a user interface reference model, which is an important aspect of the teaching of this document. Using a layered approach it can produce a complete computer system using a user interface as a key element.

SUMMARY OF THE INVENTION A resource based approach to configuration is a modelling process that enables the simplification of complex configuration problems through manageable representations which can evolve and be maintained.

Through development of a knowledge representation format and with the use of a configuration engine, many problems such as are seen telecoms can be represented in a tangible fashion. Such a technique has been applied to represent many telecoms problems. In a resource based method complex constructions are made from a number of components.

The components are defined as independent entities that are related only by the resource relationships between them. This independence provides a flexible approach in design. Two types of resource are associated with each component; the resources it consumes and the resources it supplies. For example, an electric fan supplies air flow and consumes electric power. During configuration, the requirement for airflow would be met by an instance of an electric fan. Different types of fan can supply and consume different quantities of airflow and power. Thus hierarchies of component and specialities can be constructed along object oriented principles. If the distinction was important, different attribute values can be associated with different fans. In configuration, to achieve a specific level of cooling a number of small fans may be required or a single large fan.

The choice could be based upon various criteria such as power consumption, cost of product or reliability.

Using object oriented principles, modelling is performed at the appropriate level of abstraction. Such developments may result in further refinements to the definition of the component. This can simplify problems by highlighting the important aspects. This enables the rapid prototyping of a system.

In accordance with the present invention there is provided a method of designing a telecommunications system employing a data processing means wherein products and services relating to telecommunications are represented in the form of digital data and each product and service is defined in terms of the resources it consumes and supplies, whereby the data processing means can configure a telecommunications system.

Preferably the data processor comprises a configuration engine and the products and services represented in the form of digital data comprise a knowledge base wherein an input requirement specification is determined by the configuration engine to define the telecommunications system. The present invention also provides a telecommunications design apparatus comprising a data processing means wherein products and services are represented in the form of digital data and each product and service is defined in terms of the resources it consumes and supplies, whereby the data processing means can configure a telecommunications system.

Preferably components comprising products and services are represented in a common fashion to form a knowledge base whereby an input requirement specification can be determined by a configuration engine.

By specifying a product in terms of all the possible constituent components a knowledge base is created which forms a catalogue of components and services relevant to a particular product or set of product ranges. Utilising the invention components cannot refer to other components directly, but express dependencies through the use of resources, which they may supply or consume.

By treating components as self-contained entities, the task of knowledge maintenance is vastly reduced. In addition, the inference engine to configure products expressed in accordance with the invention becomes generic, and can be applied to many different product lines and domains with significant savings. A particular configuration instance may use only a small number of the available components defined in the knowledge base. The knowledge base must be sufficiently complete to meet any requirements placed upon it but this does not prevent later additions or refinements. Once a component has been defined then reuse and evolution is both possible and practical.

The definition of components and services can easily be extracted from product design but in general the exact details usually emerge through a development process. The invention allows progressive development at the appropriate level of detail - that is, from the initial problem domain through successive developments to produce refined definitions.

Resources act at the interface between components. A resource is an entity that is stable and well defined. For example, electric current of a certain amount, or a discrete item such as a card slot of some description.

In configuration it is necessary to produce a particular instance of a product. The product instance will be an embodiment of some of the possible components, configured together in a specific way to meet specific requirements. By employing a configuration engine a given set of requirements can be achieved using a set of components. More than one solution may be possible and the choice between allowable solutions may be guided through heuristics or additional problem specific requirements.

One may desire a particular configuration of instances of components, such as a particular card in a specific slot or shelf and the configuration will be required to complete the remaining configuration problem around this constraint. In practice the configuration engine takes the initial set of instance demands and attempts to satisfy them. In doing so, a further set of demands are generated by the components to satisfy the initial demands. The process continues until no outstanding demands exist, in which case a valid configuration has been produced. The result of the configuration process is a specification detailing all of the components used and how these components are configured together. Such information can be used to cost systems networks, order equipment or to be used by field engineers in implementing the system.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a flow diagram of the steps involved in configuration; Figure 2 shows the basic relationship for a resource based model; Figure 3 is a diagram relating to the description of components for use in a configuration engine; Figure 4 is a flow diagram of the steps involved in producing an instance specification in accordance with the present invention; Figure 5 shows in greater detail the steps involved in the configuration of a system.

DESCRIPTION OF THE INVENTION Referring now to figures 1 and 2 configuration is represented as a method of arranging components and services having particular relations between the components and service. With the use of a resource based model the relationship between components and services is clearly defined. By representing both services and components in a similar fashion as digital data, the configuration engine can comprise a data processor.

Referring in particular to Figure 2, an example is shown of the basic consume/supply relationship. Components are depicted in square boxes and resources in round boxes. The notation of showing an arrow flowing from a component to a resource represents a supply relationship whilst a consume relationship is shown as an arrow flowing from the resource to the component. The component may either be of software, hardware or a feature, where a feature is a particular characteristic that the assembled artefact should exhibit. A resource can be shared if there is spare capacity. Component C could be a shelf supplying resource A, a number of slots. Components A and B are both types of card which require a slot.

So at "X" on the diagram the shelf is supplying to both cards A and B.

Component C, the shelf, also supplies another resource, resource B. The same resource B, for example power, is also supplied by component D, see 1Y'. on the diagram. Resource B is consumed by component D but neither component C or component D can individually satisfy all the demands from component B. The demand for resource B is therefore met by consuming from both component C and component D.

There may be cases where a component supplies a resource that is effctively inexhaustible. An example could be a ring software component which supplies a resource. Any piece of equipment which requires such a software function does not need a separate piece of software since one will suffice. We may consider this software as having an infinite number of software function resources. For cases such as this the digital data identification referring to this resource provides for this infinite supply.

Similarly when one component is always equipped with another component then the identification of one causes the other component to be created. This is achieved using a consume element where the resource is actually a component code. Whilst this does force the creation of a component with this code, the new component still needs all its consumes to be satisfied, and it can still supply all of its supplies.

Demands can be thought of as representing the functional requirements of the network/switch and are product independent. Demands are used to drive configuration and should capture all the information needed to produce configurations that meet the user's requirements. Demands will be in terms of traffic requirements, equipment protection and any other factors that affect the configuration of network elements. The exact specification of a network demands depends on the functional of the network elements to be configured. For example, a modular network element offers different cross-connect capabilities based on its configuration. It is important to capture the level of cross-connection required as a network demand, in order to correctly configure the network element. Network demands can simply be viewed as items that are consumed.These consumables are satisfied by resources supplied by actual physical components. Note that for the network to be configured completely all consumables must be supplied.

A component could, for example, place demands on the physical components and could themselves have demands on them by an overlaying network. Thus it can be seen that components at one level of network may form part of a demand set for components in other network levels which could in turn, place demands upon yet other levels.

The data specification can be realised using a frame based representation which provide a product code allowing data bases to access information relating to the component to be referenced. The consumes and the supplies part take the form of a list. The consumes part is used to define both the quantity and type of resources that are required by a component or feature. An optional element may be used to specify a preference ordered list when variable are used in resource names. When all of the consumable requirements are satisfied, the component or feature becomes functional, that is, it is a set of elements working together to realise a particular requirement of the final system. The supply part is used to define the types of resources that are provided by a component and their respective quantities.

Figure 3 represents a representation of two components which are defined in terms of the resources consumed and supplied.

Central to this approach to configuration is the clean separation of the configuration engine and product knowledge. The configuration engine is truly generic in nature, and the knowledge base contains only knowledge about the specific product line which is being configured. The power of this approach to configuration arises from the freedom a user may have to specify the engineering rules, or product knowledge, of telecommunications products using a simple supply-consume metaphor.

The engineer can define components and their functions. Configured systems are then constructed from the components in the knowledge base determined by the demands placed upon the system. System knowledge may be expressed so as to include: knowledge about components and their properties; relationships between components; constraints in component assembly.

For a configuration system to be simple to maintain, it is important that these kinds of knowledge are clearly identified and have an explicit representation.

Components are the building blocks of products. During configuration, these blocks are combined together to form a product configuration that provides the user with a specified function. Component knowledge defines the properties of each building block, such as its name, product code, unit cost and description. This knowledge is often used to support the task of selecting configuration options, such as when two components satisfy a particular demand.

Some component knowledge, such as component name or component code, will always be expressed within the knowledge base. Other attributes of components, such as size or weight can be expressed in an external database, but it is also possible to encode this non-functional knowledge within the knowledge base.

Relationships between components are expressed in terms of the resources they supply and the ones they consume. These relationships form a chain of dependencies between components, which for many telecommunication products map onto cards providing specific features, which in turn require common cards, such as power cards, which together have a requirement for a rack, and so on.

This explicit representation of the resources that each component supplies and consumes has the benefit of providing an elegant way of expressing the sharing of resources by components, for example, the sharing of the current provided by a power card.

It is important to note that a component contains no direct references to other components, it only refers to resources. This is because a component does not need to know what is supplying the resource, only that the resource is supplied in order to satisfy the component's demand.

The components can thus be considered in isolation.

A constraint is a restriction associated with a component which affects its supply-consume reltionships. Constraints typically express electrical or spatial restrictions on card placement. Constraints are specified in terms of resource values. Spatial constraints may express relationships such as "Power Cards can only go in slot 14 because they are too big", or "Card A and Card B cannot be placed next to each other because of their size".

Electrical constraints can express relationships such as "Card A can consume current from a Power Card only if it is on the same shelf as that Power Card".

The purpose of the configuration engine is to generate assemblies of components (which components are either products or services) through the product specification data. A key to the configuration engine is the representation of configured or partly configured components in the system. In a data processor system generating a design these shall be referred to as factual instances - see Figures 4 and 5. These factual instances are identified by unique identifier for each component instance, the code of the component, the supplied resources (type of resource, quantity and identification of this supplied entity) and the consumed resources (type of resource, quantity and an identification of the supplying entity).The instance demands or requirements are similarly represented in that they are identified by the type of resource required, the quantity and the entity requesting the requirements.

The configuration engine configures a system taking into account the outstanding demands or requirements, a list of component instance terms describing components already included in the configuration and a list of configured components which represent an assembly of components which together satisfy the demands specified by the outstanding demand.

The configuration engine recursively deals with each demand in turn until there are no outstanding demands. The operation of the configuration engine may add demands to the outstanding demands when new components are created as well as removing them when they are satisfied. In operation of the configuration engine, when the outstanding demands list is empty, the configuration is complete. Thus configuration can be optimised to produce a system with as few components as necessary.

Accordingly the configuration engine first checks whether any of the preconfigured components have the required resource to offer. Once it has been ascertained that an existing component supplies the required resource then a determination of the quantities involved is reckoned. If the quantity specified in the demand matches the quantity supplied by the preconfigured component then the supply term is removed from the supplies list of the preconfigured component and converted to a supplied term which is added to its supplied list, the original component instance term is replaced by this new description in the preconfigured components list to form a new list. Resource arithmetic may be employed to evaluate arithmetic operations on numeric resource values.

If the preconfigured component supplies more of the resource than is specified in the demand then the supply term is removed from the supply list of the preconfigured component and replaced with the supply term for the remainder of the resource. A supplied term is added to the component's supplied list specifying the amount that was consumed.

Again the original component is replaced by this new component description.

If the demand specifies a quantity greater than that supplied by the preconfigured component, a similar manipulation of the component is then effected such that the required resource is supplied. Since it cannot supply the entire quantity the remainder of the demand is added to the front of the other remaining outstanding demands list. If there are no existing components which can satisfy the current demand, or if the reuse of existing components led to a failure, the configuration engine must create a new component.

In creating a new component the configuration engine will first look for a component class which can supply a resource of the required type. If an appropriate class is found then a component instance fact is asserted with a further unique identification. The supplied list parts of this component type are generated from the supplies list of the class and the current demand, much in the same way as the component reuse example above.

The consumables list of the class is converted into a list of demand terms and if no component classes could supply resource which matches the current demand it may be that the demand represents an included component. This is the case if the demanded resource matches the code argument of a component class. A number of component instances are then created to gather up all the demands generated by the components.

The present invention provides a powerful and expressive mechanism for defining components and their functions using resources. The main benefits are that product knowledge is kept separate from the generic configuration engine, which ensures that there is no embedded problem solving knowledge; maintenace can be easily performed as components are self-contained entities, with no direct reference to other components; and the wide ranging applicability of the approach to different domains.

As can be envisaged, the prior reiterative procedure can be time consuming. By representing both products and services as digital data in the configuration of a telecommunications system then the design of such a system can be expedited. A significant benefit of using the present invention is that there are no implicit or explicit references to other components, as components only provide and make use of resources.

This leads to easier maintenance as components may be treated as black boxes, as a change in one component will have no impact on another.

Claims (4)

1. A method of designing a telecommunications system employing a data processing means wherein products and services relating to telecommunications are represented in the form of digital data and each product and service is defined in terms of the resources it consumes and supplies, whereby the data processing means can configure a telecommunications system.
2. A method according to claim 1 wherein the data processor comprises a configuration engine and the products and services represented in the form of digital data comprise a knowledge base wherein an input requirement specification is determined by the configuration to define the telecommunications system.
3. A telecommunications design apparatus comprising a data processing means wherein products and services are represented in the form of digital data and each product and service is defined in terms of the resources it consumes and supplies, whereby the data processing means can configure a telecommunications system.
4. An apparatus according to claim 3 wherein components comprising products and services are represented in common fashion to form a knowledge base whereby an input requirement specification can be determined by a configuration engine.
GB9421227A 1994-10-21 1994-10-21 Communications system design. Withdrawn GB2294340A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
GB9421227A GB2294340A (en) 1994-10-21 1994-10-21 Communications system design.

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
GB9421227A GB2294340A (en) 1994-10-21 1994-10-21 Communications system design.

Publications (2)

Publication Number Publication Date
GB9421227D0 GB9421227D0 (en) 1994-12-07
GB2294340A true GB2294340A (en) 1996-04-24

Family

ID=10763179

Family Applications (1)

Application Number Title Priority Date Filing Date
GB9421227A Withdrawn GB2294340A (en) 1994-10-21 1994-10-21 Communications system design.

Country Status (1)

Country Link
GB (1) GB2294340A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1998048364A1 (en) * 1997-04-18 1998-10-29 Network Tools Computer-aided-design method and apparatus for networks
DE19914216A1 (en) * 1999-03-29 2000-10-12 Siemens Ag Automatic assignment of addresses in transmit and receive buffers
US6272390B1 (en) 1997-09-09 2001-08-07 Beologic A/S Method for interactive configuration and computer readable medium encoded with a program for performing said method
US7096165B2 (en) 2000-03-17 2006-08-22 Siemens Aktiengesellschaft Method for configuring an electrical installation and corresponding configuration device

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1998048364A1 (en) * 1997-04-18 1998-10-29 Network Tools Computer-aided-design method and apparatus for networks
US6058262A (en) * 1997-04-18 2000-05-02 Geargarage.Com Inc. Computer-aided-design method and apparatus for networks
US6272390B1 (en) 1997-09-09 2001-08-07 Beologic A/S Method for interactive configuration and computer readable medium encoded with a program for performing said method
DE19914216A1 (en) * 1999-03-29 2000-10-12 Siemens Ag Automatic assignment of addresses in transmit and receive buffers
US7096165B2 (en) 2000-03-17 2006-08-22 Siemens Aktiengesellschaft Method for configuring an electrical installation and corresponding configuration device

Also Published As

Publication number Publication date
GB9421227D0 (en) 1994-12-07

Similar Documents

Publication Publication Date Title
Bernus et al. Architectures of information systems
van der Aalst et al. Business process redesign: A Petri-net-based approach
Kosanke CIMOSA—overview and status
Kang et al. FORM: A feature-; oriented reuse method with domain-; specific reference architectures
US6229540B1 (en) Auditing networks
Griss et al. Integrating feature modeling with the RSEB
Wies Using a classification of management policies for policy specification and policy transformation
CA2207867C (en) Method and apparatus for providing an efficient use of telecommunication network resources
Olsen Systems engineering using SDL-92
US6275977B1 (en) Application cooperation method and apparatus
US20080140676A1 (en) Supply chain/workflow services in a contract manufacturing framework
Wang et al. QoS-enabled middleware
Damianou et al. Tools for domain-based policy management of distributed systems
US6237036B1 (en) Method and device for generating access-control lists
US6330005B1 (en) Communication protocol binding in a computer system for designing networks
Govorskii et al. Mathematical modeling of inhomogeneous traffic in a heterogeneous integrated corporate data control system
Kim et al. A modeling framework for agile and interoperable virtual enterprises
Burrafato et al. Designing a multi-agent solution for a bookstore with the PASSI methodology.
Christensen et al. Coloured Petri nets extended with channels for synchronous communication
US20070211280A1 (en) Method and apparatus for assigning candidate processing nodes in a stream-oriented computer system
US6430614B1 (en) Method and apparatus for providing and facilitating interaction with distributed manager information of a network
US6058262A (en) Computer-aided-design method and apparatus for networks
Bosch Maturity and evolution in software product lines: Approaches, artefacts and organization
JP2010512079A (en) Apparatus and method for network management
US5831610A (en) Designing networks

Legal Events

Date Code Title Description
WAP Application withdrawn, taken to be withdrawn or refused ** after publication under section 16(1)