Abstract: A method and system for metering usage of software products on a data processing system. Requests to load an executable module are monitored and, if it is determined that a request was received to load an executable module, a metering flag is checked for the executable module that is indicative of whether the executable module is an executable module that is to be metered. If the metering flag indicates that the executable module is to be metered, a software product associated with the executable module is identified.

Abstract: A method (300) for metering usage of software products in a data processing system is proposed. The products to be metered are identified in a global catalogue (325) by means of multiple attributes of a corresponding executable module (340); the verification of these attributes requires accessing the module through its path, which is not available at run-time (being the module directly identified by a corresponding location for performance reasons) . The solution is based on the idea of creating a local catalogue (350) that associates the path of each module installed on the system (that is included in the global catalogue) with its location. The locations of the running modules are detected (365) periodically from a list of processes in execution on the system. In this way, it is possible to obtain (355) the path corresponding to the location of each running module in a very fast way trough the local catalogue (without any scanning of the system).

Abstract: A method and a corresponding apparatus for metering usage of software products on a computer are proposed. The metering system is based on the use of a software catalogue. The software catalogue lists all the known products to be metered; each product is identified by one or more executable modules indicative of its running. A licensing agent detects the executable modules that have been launched. For each executable module, the system checks if a corresponding product is listed in the catalogue, and can decide an appropriate action, or simply meter the usage of the product for future reporting and invoicing activity. If no corresponding product is found in the catalogue, the method according to the present invention performs a similarity check between the unknown detected module and the product identifiers existing on the catalogue. Sometimes the difference between a new product and a catalogued one is very limited.

Abstract: A method and a corresponding apparatus (200) for metering usage of software products on a computer are proposed. The solution of the invention is based on the idea of associating (225) each product with an installation signature (indicative of the installation of the product on the computer) and with a running signature (indicative of the running of the product on the computer). The products that are installed on the computer are asynchronously determined (235) according to their installation signatures; in this way, any ambiguities (for example, caused by different versions and/or configurations of a product sharing the same executable modules) can be resolved in advance. Therefore, when at run-time the invocation of an executable module is detected (245) it is possible to use only the running signatures to identify the corresponding product uniquely (without the need of additional information about the executable module, such as its size).

Abstract: A method and a corresponding system (200) for metering execution of interpreted programs (220) are proposed. Typically, an interpreter, such as the Java Virtual Machine (215), lacks built-in capabilities for communicating with the hardware and software platform of the computer directly; for this purpose, the virtual machine invokes external native functions through the Java Native Interface (225). The solution of the invention is based on the idea of stubbing the dynamic libraries (235) that implement those native functions. Therefore, when the virtual machine needs a native function (for example, to load a new class), a stub library (250) is loaded in place of the required dynamic library. As soon as the virtual machine calls the desired command, the stub library takes control and identifies the interpreted program that is run by the virtual machine; preferably, this operation is performed by querying a catalogue that associates the new class with the corresponding interpreted program.

Abstract: A method (300) and system for controlling use of software programs on multiple workstations of a network. An agent running on each workstation detects (306) any new process (using a kernel hooking technique), which is started in response to the request of execution of a program (397). The process is suspended (308) and an authorisation to run the program is requested (334) to an associated metering server. The metering server compares (380) the current usage of the product with the respective authorised conditions of use, and answers (382) the agent granting or denying the authorisation to run. The process is accordingly resumed (312), in order to enable the program to start, or aborted (315), so as to prevent execution of the program. In any case, the program is not stopped (but it is prevented from starting); this avoids killing the program while it is performing potentially risky operations.