International Business Machines Corporation

Vincenzo Sciacca

Massimo Villani

Francesco Carusi

Rosario Gangemi

Attilio Mattiocco

Scot MacLellan

Scot Maclellan

A resource consumption model is created for a software application, making it possible to predict the resource requirements of the application in different states. The model has a structure corresponding to that of the application itself, and is interpreted to some degree in parallel with the application, but each part of the model is interpreted in less time than it takes to complete the corresponding part of the application, so that resource requirement predictions are available in advance. The model may be interpreted in a look-ahead mode, wherein different possible branches of the model are interpreted so as to obtain resource requirement predictions for the application after completion of the present step. The model may be derived automatically from the application at design or compilation, and populated by measuring the requirements of the application in response to test scenarios in a controlled environment.

Predicting resource requirements for a computer application

Embodiments are disclosed for managing a plurality of virtual computers in a virtual environment. According to one embodiment, a method includes creating a virtual group. The virtual group includes a plurality of virtual computers having a relevant characteristic. An action to perform on the virtual group is received, and the action is performed on each of the virtual computers in the virtual group.

Managing virtual computers simultaneously with static and dynamic dependencies

A method for monitoring transaction based systems monitoring their system resource and transaction performance data with thresholds on metrics, said method comprising the steps of collecting system resource data and transaction performance data on a period basis and each time a transaction performance metric is passed, triggering the collection of the system resource and storing an identifier correlating the transaction performance data and the system resource data; merging the collected system resource and transaction performance data; calculating the mean value for each system resource or transaction performance metric of merged data; identifying the metrics for which there is a significant difference between mean value obtained with triggering or without triggering; according to the identified metric mean value, calculating new thresholds of system resource metrics to be used for monitoring.

Method and system for monitoring transaction based system

A solution for distributing the workload across the servers () in a fail-over cluster (for example, based on the MSCS) is proposed. A fail-over cluster is aimed at providing high availability; for this purpose, a resource service () automatically moves each resource () that exhibits some sort of failure to another server in the cluster. The proposed solution adds a monitor () that periodically measures a responsiveness of each resource. If the responsiveness of a resource is lower than a threshold value, the monitor inquiries a metrics provider () for determining the workload of all the servers in the cluster. The monitor then causes the resource service to move that resource to the server having the lowest workload in the cluster.

Fail-over cluster with load-balancing capability

A method and system for electricity consumption profile management for electrical devices, is provided. One implementation involves receiving an energy consumption control command at a consumer site including on or more electrical devices connected to an energy distribution network; determining energy consumption adjustment including determining an energy consumption adjustment for at least one of the electrical devices based on the energy consumption control command; and transmitting an energy consumption adjustment message to said at least one of the electrical devices, commanding the electrical device to adjust its electrical energy consumption accordingly.

Method and system for electricity consumption profile management for consumer devices

Method and system for performance monitoring of distributed applications. A sensor intercepts every request of service for a server that is generated on a client. If the request meets a filtering condition (for example, defined by the address of the server, the web page from which the request is originated and/or the selected link) the measuring of a corresponding transaction on the client is enabled; at the same time, the request is updated by inserting a correlator. The request is then transmitted to the server. If the request includes the correlator, the measuring of a sub-transaction originating from the request is also enabled on the server. The parameters measured on the client and on the server are then associated with the correlator.

Method and system for monitoring performance of distributed applications

A method and system for instantiating COM components in custom contexts are proposed. The solution of the invention is based on an auxiliary component, acting as a factory (). Whenever a client () needs to instantiate a new target component in a desired custom context, it calls a dedicated method exposed by the factory, passing the identifier of the target component together with information about the custom context. The factory starts execution of an additional module, operating as an impersonator (), in a new process () conforming to the custom context. The impersonator then instantiates the target component in its own process. The impersonator passes a pointer for the target component to the factory (calling a further method that is exposed by the factory); the factory in turn returns that pointer to the client.

Growing community of inventors

Rome, Italy

Vincenzo Sciacca