Patent Application
20130013614
The present invention relates to identifying software. More particularly, the present invention relates to software license management, identifying usage of software and management of software on a network.
A network of computers can be very large with hundreds or more computers on the network each with one or more software packages. Various computers may have different versions of the same software. Many software packages are subject to licenses and may be also subject to maintenance charges. Managers wish to manage a network in a cost efficient manner. It is necessary to ensure all software which requires a license is properly licensed. Also it is desirable that unused software is removed to reduce the number of licenses and maintenance agreements to reduce costs. Systems for doing that have been proposed.
It is desirable to minimise the number of different versions of what is nominally the same software to minimise support and administration costs.
It may also be desirable to replace one version of software by another version to reduce costs and/or to provide the user with software more appropriate to their needs.
In order to manage a network, and for the purpose of managing licenses, it is necessary to identify the software installed on the network. One aspect of the present invention seeks to provide a process for reliably identifying software installed on a computer, or on computers in a network.
The present invention is based on the recognition of an unexpected problem. It is conventionally expected that software names conform to the pattern publisher, product name, version identifier, for example “Microsoft Office Standard 2010 10.0” (Microsoft is a Registered Trademark).
However there is in fact no standard so some software names do not conform to that pattern. Furthermore, what is the same software may be marketed under different names, and/or version identifiers (also known as a release numbers). Also, it has been found that programmers, or others responsible for naming software, make mistakes such as representing the publisher's name incorrectly. This creates difficulties in identifying software and determining what software packages are in the fact the same.
In accordance with one aspect of the present invention, there is provided a computer implemented method of establishing the identity of software installed on a computer, the software being purportedly identified by identification data associated with the installed software, the method using a catalogue of established software identification data, the method comprising:
By using such mapping, a coherent and consistent set of software identification data is created so that for example plural items of the same software are identified by the same established identification data. That simplifies the administration of software.
The method may be applied to a computer network in which the method is carried out by an identification processor which accesses the data base of a systems management tool.
Another aspect of the invention provides a computer system comprising a plurality of user's computers and one or more administrative computers linked by a network, the system comprising:
Thus, examples of the present invention are based on the recognition that software installed on a computer network is often inconsistently, or even incorrectly, identified. The same software may be identified in different ways. A catalogue of standardised identifiers is provided. The actual identifiers of software installed on the network are accessed and they are mapped to the standardised identifiers of the catalogue. The standardised identifiers are used to manage the installed software, monitor license compliance and/or, monitor maintenance agreements amongst other uses. Data relating to the use of the software may also be obtained and associated with the identification data. The usage data together with the standardised identifiers allows managers to more reliably manage software on the network. For example un-used software may be un-installed and licenses cancelled or reallocated.
Embodiments of the present invention use identification data as provided by software authors and publishers in the software even though that data may be subject to errors and inconsistencies. The identification data used is the conventional publisher name, and/or product name together with the version identifier if it exists. Embodiments of the present invention also use the conventional product codes which are or should be in the software if they are present. Thus, normal identification data, as conventionally provided by authors or publishers, are used without the need for special or unconventional identification codes.
Further features and advantages of the invention will become apparent from the following description of illustrative embodiments of the invention, given by way of example only, which is made with reference to the accompanying drawings.
An aim of embodiments of the invention is to:
i) discover what programs, e.g. software applications, are installed on one or more computers.
Embodiments of the invention may also
ii) monitor usage of the programs;
iii) discover what different versions of the nominally same software are being used; and/or
iv) reconcile the discovered programs with a database of licenses.
That enables a network manager to: ensure software is appropriately licensed; move all users to the same and most recent version; save costs by reducing licence administration of many different versions; reduce the cost of supporting many different versions of software; save costs by eliminating unused software; eliminating or reallocating unused licenses; and/or provide software appropriate to the needs of the users.
Software is apparently conventionally identified by publisher name, product name and version identification: e.g. “Microsoft Access 2010 10.2”. It would at first sight seem simple to derive the identity of software packages from such identification but it is not because in fact there is no standard way of identifying software which all providers adhere to. Even one publisher may not identify their own software packages in a consistent way. Also what is basically the same software may have very many different versions, or different items of the same version may be identified in different ways. Generally it has been found that authors of software make errors in identification, make spelling errors, do not include all the version information, and/or do not adhere to the apparently conventional “publisher name, product name, version identifier” format. For example it has been found that in some cases the version identifier has been combined with the product name or the publisher name has been combined with the product name. Such identifiers are described as “noisy” hereinafter. In some cases the product name and the publisher name are the same; e.g. winzip.
The following description refers to databases. The term “database” is used in the general sense of a collection of data stored in a data store. A database described herein by way of ease of description as a single database may be a set of different data tables from which the relevant data is derived.
Embodiments of the invention will be described by way of example with reference to a computer network as shown in
Users of the network have computers 2, which may be on subnets 4, and are linked by the network symbolised at 6. Some computers 8 may be roaming away from the network 6. For example, the network may be linked to another network 7, for example the Internet, and computers 8 owned or managed by the enterprise may roam and use the Internet to connect to the enterprise network or connect to it in other ways, for example via a Virtual Private Network (VPN) 7.
The network comprises a systems management tool (SMT) 10. An example of such a tool is a Configuration Manager (ConfigMgr), also formerly known as Systems Management Server (SMS), provided by Microsoft Corporation. Other companies also provide systems management tools. The tool 10 has a database 101 storing data relating to the configuration of the network. The systems management tool uses agents on the computers 2, 8 which report regularly to the systems management tool 10. An agent A on a computer 2, 8, identifies the software installed on the computer and the times of use of the software and reports regularly to the systems management tool 10.
An installer, for example Microsoft Installer, 10 may be used for installing software on computers of the network or software may be installed by users directly on their computers. The agents detect the software however it is installed.
The term “software” includes patches for, and updates of, software already installed on the network, and new software packages. The database 101 stores data identifying all software installed on computers of the network using the systems management tool.
The network also has a license manager 14 and database 141 for recording data relating to the licensing of software installed on the network. The data of the license database is entered by the user via an interface 142. As will be discussed, the license manager may also store in the database 141 data relating to maintenance agreements for installed software. That data also would be entered by the user via the interface 142.
The network also has an identification and usage processor 16 which accesses a database 161 storing a catalogue of established software identifiers. In the example of
The catalogue also includes for each software item an indication of whether it should be licensed. The catalogue may also include product codes but the catalogue may in practice be only sparsely populated with product codes.
Referring to
An example of the mapping process uses an English dictionary. Referring to
Referring to
Referring to
Machine 1, software ID1, software ID2 . . . .
Machine 2, software ID1, Software ID5 . . . .
Usage data, which will be described hereinbelow, is also stored. The usage data is reported by the agents A.
The process of
The installation data (i.e. the identifiers of the installed software) is “standardized” or “sanitized” as indicated at 77 by a process which will be described with reference to
As indicated at 78, license information including the identifiers of licensed (and of maintained) software are entered into the licensing database 141. The entering of the identifiers is done manually via the interface 142 in this example. The user enters the license information based on their records of license agreements and any maintenance agreements. Those identifiers may be “noisy”. As indicated at 80, the identifiers are “standardized” by the process, using the catalogue and dictionary, which will be described with reference to
As indicated at 83, usage data relating to the usage of the software installed on the computers is obtained from the systems management tool 10 and stored locally. The agents A on the computers of the network report the usage of the software installed on them to the systems management tool. The usage data is preferably standardized as indicated at 84 by the process which involves reference to the catalogue 161 and the dictionary 162 of
The usage data has the form by way of simplified example:
In comparison to the usage data, the corresponding installation data has the form by way of simplified example
The standardized usage data is processed as indicated at 85 by a process which will be described with reference to
As indicated at 87, the software usage and identification data associated with each one of the computers 2, 8 is checked to determine if it is trustworthy. Whilst
As indicated at 88, the standardized identification data and the processed usage data are correlated with the standardized license and maintenance data and the correlated data used as indicated at 89 as will be described later.
One embodiment of the mapping process is described with reference to
There are three data items interest and each is processed in a different way: Publisher, Product, version identifier.
Examples of noisy data are:
There are two real products here which appear in the catalogue as:
The year number, in the examples above “2010”, is deemed to be part of the version identifier in the catalogue. It may be omitted from some software identifiers.
Referring to
Step S41 starts the process by finding the publisher data.
Step S42 splits the publisher field into individual works. For example, if the publisher data is Microsoft plc it is split into Microsoft and plc.
Step 43 removes words like plc or GMBH or any other legally descriptive words. That may be done by creating a list of such words, comparing the words of the publisher data with the list and removing any word which matches any word in the list.
So step S43 produces
Step S44 compares the word(s) of the publisher name with the dictionary. It is expected that names such as “Microsoft” will not appear in the dictionary but other words, e.g. “system” will be in the dictionary.
Step S45 applies a weighting to each word of the publisher's name: for example 1 if is NOT in the dictionary and 0 if it is. The assumption here is that publisher names are not in the dictionary. Words such as “apple”, and “adobe”, which are also words used in publisher's names, are exceptions to this and are dealt with as described below.
Step S46 attempts to match the word(s) of the publisher remaining after step S43 with words in the catalogue. An illustrative matching process is described with reference to
As shown at step S461, if the there is only a single word of publisher name which matches a word in the catalogue, the matching catalogue name is taken as the established publisher name whether the word is weighted 1 or 0. For example, if the single word is “Microsoft”, that is taken as the established publisher name.
As shown at step S462, if there are two words in the publisher name, e.g. Microsoft Systems, and one (e.g. “Microsoft”) is weighted 1 and the other (“systems”) is weighted 0 then the word weighted 1 (Microsoft) is taken as the publisher name because it has the greater weight because it is not in the dictionary.
As shown at step S463, if there are for example two words both NOT in the dictionary (i.e. both have weight=0) but both words are in the catalogue, (for example Microsoft Softsys) then the word occurring most often as publisher in the catalogue is chosen as the established publisher name.
So, at the end of step S461, the software identifiers are changed to
The publisher name has been matched to the catalogue.
An exceptional case is where you have an English word such as “apple” as a publisher name. For example:
In these examples all the words are in the dictionary. As shown at step S461, if there is only a single word, e.g. Apple, step S461 attempts to match it to the catalogue (in this case Apple is in the catalogue). If there are two or more words all in the dictionary and in the catalogue (e.g. Apple Systems), as set out in S463, the matching selects the lexicographically lowest entry, again Apple in this case.
As set out in S464, if for example the word Apple is in the catalogue but Systems, Division or Eastern is not in the catalogue, Apple is selected as the publisher name.
Step S50 finds the product field. The inventor has found that the product field often contains the publisher name. Step S51 removes the publisher unless it is the sole content of the product field: “winzip” is an example where the publisher name and the product name are the same. Data may be identified as a publisher name by comparing the data with publisher names in the catalogue. The result of step S51 is:
“Microsoft” is gone from the product name of example 4.
Step S52 processes the remaining information via a number of regular expressions. This removes characters, dates and other data in a list of regular expressions.
Step S53 attempts to find anything looking like a version identifier. Step 53 in this example uses heuristics that find version identifier by pattern matching. For example a heuristic detects: dots; v followed by a number; and other indicators of version identifiers. Step S53 removes the version identifiers but also stores the removed version identifiers for use later. Step S53 produces
Step S54 performs a catalogue match against the products of the publisher established in the process of
The result of step S54 is
Thus the product name is established.
Processing the Version Identifiers
Processing the version identifiers aims to provide a name (for example a year number) and major and minor release numbers.
Step S53 of
Step S60 extracts the version identifier (if any) from the version field of the data obtained from the SMT database 101.
The version identifier (if it exists), whether derived from step S53 or from a non-empty version field, is processed in step S61 using heuristics attempting to extract a name (e.g. a year number if it exists and major and minor release numbers.
For example referring to the examples of identifiers again,
Step S62 adds any data stored in step S53 and processed by step S61 to the publisher and product names established from the catalogue. It also adds any the version identifier derived from a non-blank version field and processed by step S61.
For example, the result of step S62 is
Step S63 matches the publisher and product names to names in the catalogue and then tries to find the best version data in the set of catalogue data of the matched publisher and product names. For example, for
line 2 has a version name (2010) and they all have matching major and minor release numbers so we can also use the version name for 1 and 2 resulting at S64 in the established data.
The same applies to the Office standard entries and steps S63 and S64 provide:
Variations
Sometimes the publisher field is missing but the publisher data appears in the product field. If so the publisher data is taken from the product field and put into the publisher field.
This may be achieved by determining if the publisher field is blank and then examining the product field. The product field in then processed as set out in steps S42 to S46 in
Some publisher names are also dictionary words: e.g. adobe. They can be found by matching words in the product field with publisher names in the catalogue.
If the publisher data is missing, the product and version as processed by the processes of
Result of Processing
The result of the processing of
Product Codes
The installation database 163 also includes the product codes of the identified software. The product codes may be GUIDs. The product codes are derived from data in the software. Authors of software should include product codes but in practice they are sometimes omitted. The catalogue may also include product codes but may be only sparsely populated with product codes or they may be absent,
Processing Usage Data—
Usage data for software installed on each computer on the network is stored in the database 101 of the Systems Management Tool 10. Collection of such usage data is a known function of a Systems Management Tool which uses the agents A on the computers to collect the usage data.
The usage data has the form by way of simplified example:
The corresponding installation data has the form by way of simplified example
The processing of the usage data attempts to match the usage data to the standardised software identification data in the installation database. The processing of the usage data aims to identify the usage of software installed on the individual computers and to add, or otherwise link, the usage data to the standardised identification data in the installation database 163.
The usage data derived from the installed computers includes product codes.
Referring to
The stored data is standardized as indicated at 84 using the processes of
For each item of software installed on a particular computer the following occurs.
Step 90 determines if a product code is in the standardized usage data. If yes, then step 91 attempts to match the product code in the usage data
with a product code in the installation data (i.e. the standardised software identification data). If the product code matches with one in the installation data, the installation data is associated with the usage data as indicted at 86.
If step 90 does not find a product code in the usage data, step 92 attempts to infer a product code. A product code is inferred by matching the publisher name-product name-version identifier-tuple of the software associated with the usage data with that of other software which has a product code. If there are multiple product codes which all group to the same publisher-product-version-tuple, then all those product codes are considered as a bundle. The bundle is then used to match up corresponding data relating to software installed on the particular computer. The usage data is then associated at step 86 with the installation data which can be matched to the bundle of product codes.
If there is no standardising step 84 and if no match can be achieved using product codes (whether inferred or not), then the step 94 standardises the usage data.
Step 95 attempts to match the usage data to the installation data using the sanitized publisher name-product name-version identifier-tuples in the usage data and the installation data. If a match is found, the installation data is associated with the usage data: if not step 96 is carried out.
In step 96, the usage data is compared with the catalogue to find a match. Also installation data is compared with the catalogue to find a match. If installation data and usage data match the same data in the catalogue, then the usage data is deemed to be associated with the installation data.
If product codes are present in the catalogue, usage data may be linked to installation data using the product codes and the catalogue.
Result of
The processing of
Network Management
The data produced by the processing of
The license database 141 contains data which has been standardised as indicated at 78, 80 and 82 in
One embodiment of the invention aim to monitor licence compliance. The embodiment finds out what how many packages of a particular software are installed, finds out how many licenses have been bought, and then indicates compliance, or not compliance, with the terms of the licenses. If there are more licences than needed some maybe cancelled. If there are insufficient licenses more may be bought. The embodiment may automatically generate a report.
Another embodiment uses the usage data to aid in deciding whether to uninstall software which is not used and/or re-allocate any associated license. The embodiment may automatically generate a report.
Another embodiment uses the usage data to determine if installed software is inappropriate to any user. The usage data may be used to determine if other software is more appropriate to any user(s).
Another embodiment determines whether various different versions or releases of the same software are present on the network. The embodiment may automatically generate a report. The various versions may be replaced by a single version, which may assist in reducing support costs
Another embodiment considers the combination of a product license and a maintenance contract: the combination is termed herein the “effective license”. The maintenance contract may not coincide in time with the product license: for example maintenance may need renewal annually whilst the product license may be a one-off cost. The embodiment aims to monitor the effective license. Is there for example a gap in the maintenance: has the maintenance lapsed? The embodiment provides a rules engine to determine whether product license is currently valid and whether the maintenance is up to date and to produce a report on compliance. A manager can decide to reallocate licences, and/or uninstall software, and/or buy more maintenance or stop maintenance.
Such embodiments depend on reliable identification of the installed software. The processes of
Computer Programs and Program Carriers.
Embodiments of the invention may be implemented by a program or a set of programs which, when run on a computer or set of computers, causes the computer(s) to implement the methods described herein above. In one illustrative implementation of the invention the programs include software implementing the identification processing as indicated in
Software may be provided to correlate the identified installed software and usage data with data in a licensing data base as indicated at 88 in
Software may be provided to aid network management as described above.
The program or programs may be carried by one or more carriers. A carrier may be a signal, a communications channel, a non-transitory medium, or a computer readable medium amongst other examples. A computer readable medium may be: a tape; a disc for example a CD or DVD; a hard disc; an electronic memory; or any other suitable data storage medium. The electronic memory may be a ROM, a RAM, Flash memory or any other suitable electronic memory device whether volatile or non-volatile.
The above embodiments are to be understood as illustrative examples of the invention. Further embodiments of the invention are envisaged. Whilst
