Abstract: A method includes, in a grid network, maintaining a store containing a history of specific application executions, maintaining a list of grid applications in parallel, executing an application capable of executing in parallel on one or more grid compute nodes contained in the list, terminating execution of the application on the one or more grid compute nodes upon a successful completion of the application in any of the one or more grid compute nodes, and transmitting a result to a client.

Abstract: A method includes, in a grid network, maintaining a list of applications that can solve a task, the list referencing at least one primary application and a plurality of secondary applications used to solve the task, executing the task with the primary application in response to a request to solve the task, and executing the task with one or more of the secondary applications according to grid resource availability.

Abstract: A system includes a first device that runs an archiving master program. The first device includes available storage for use by the archiving master program. A second device runs a computer program that generates log files for storage on the second device. The second device runs an archiving slave program. The archiving slave program sends the log files to the archiving master program. The archiving master program receives the log files from the archiving slave program and stores the log files in the available storage.

Abstract: A method includes, in a network of interconnected grid compute nodes, receiving a request to execute an application in the network, registering the application, determining whether the application meets a threshold to enable assigning the application to one of the grid compute nodes within the network, assigning the application to execute on a specific grid compute node having no other current applications executing, preparing the grid compute node for execution of the application, recovering the grid compute node if the application terminates prematurely, and deregistering the application on the grid compute node if the application executes successfully.

Abstract: Methods and apparatus, including computer program products, for control interfaces for distributed system applications. A method includes, at an application deployed on a first computing system in a grid computing environment, monitoring a communication channel connecting the first computing system to a second computing system, receiving a command over the communication channel from the second computing system, and in response to the received command, generating a descriptor file including descriptions of one or more actions to be performed by the second computing system in order to move the application from the first computing system to a third computing system.

Abstract: A system is disclosed for obtaining debug information in a network that supports Web services. The system includes a first service host to send a first message, where the first message contains data corresponding to a request for debug information. The system also includes a second service host to receive the first message and to send a second message to a third service host in response to the first message, where the second message contains data corresponding to a request for debug information. The second service host receives a reply to the second message from the third service host, where the reply to the second message includes third debug information associated with the third service host, and the second service host generates a reply to the first message. The reply to the first message includes the third debug information and second debug information associated with the second service host.

Abstract: A method includes identifying a component within a non-distributed grid-enabled software application that can be isolated and independently executed during an execution of the software application on a first compute resource within a grid network, replacing the isolatable component in the software application with a proxy, and executing the software application on the first compute resource with the isolated component executing on a second compute resource.

Abstract: A method includes, in a network of interconnected computer systems, storing a first list of links represented as universal resource locators (URLs) that match a first search request, displaying a first page in response to a navigational input, highlighting links on a second displayed page that match links contained in the stored first list.

Abstract: Methods and apparatus, including computer program products, for ad-hoc solution resources determination and provisioning. A method includes, in a target compute system residing in a network of interconnected compute systems, assigning a unique identifier to a requested solution, selecting providers for providing identified components of the requested solution, storing identifications of the selected providers along with associated identified components in a hierarchical data structure in a memory of the target compute system, each of the associated identified components defining a set of required resources, and installing the identified components listed in the hierarchical data structure.

Abstract: Methods and apparatus, including computer program products, for matching software requirements against target system landscape descriptions and for applying rating metrics to intermediate results during the matchmaking process. Data are received as inputs describing the requirements and dependencies of a particular software application and the resources available in the target system that may be able to meet these requirements. These data are searched for combinations of system resources that will satisfy the requirements of the application as well as optimization metrics and other requirements supplied by the user. Once a match is found, it is given a rating and ranked against any other matches.

Abstract: A method in a network of interconnected computer systems includes generating a Quality of Service (QoS) requirements document between a client and a first server for execution of a service in the first server, initiating execution of the service in the first server in response to a client request, and migrating execution of the service from the first server to a service in a second server if the first server cannot guarantee requirements specified in the QoS requirements document.

Abstract: A method that includes, in a network, in each of a plurality of compute nodes, maintaining a local store of services, the local store of services including at least a service name, a service functionality and statistics defining a historical performance of the service on each compute node, receiving a request for a service from a client syste, and compiling a list of compute nodes matching the service request, the list having service names ranked according to the service functionality and the statistics for each compute node.

Abstract: A system and a method includes generating customization settings for an application on a first grid system, the customization settings enabling grid operation of the application. The system and method also includes distributing the application along with the customization settings to a second grid system, and installing the application and the customization settings on the second grid system such that computational resources of the second grid system can be allocated to the application from another grid system.

Abstract: A method including, in a grid network, determining available compute devices for installation and execution of an application, the application requiring a specific operating system. The method also includes, in response to determining that there are no available compute devices having the specific operating system, sending a request to install the specific operating system and a grid manager through a link to a management system. The method also includes installing the specific operating system on one of a plurality of compute devices controlled by the management system and linking the one of a plurality of compute devices controlled by the management system to the grid network. Lastly, the method includes installing the application on the one of a plurality of compute devices controlled by the management system.

Abstract: A system and a method includes storing one or more configuration element entries in a definition scheme on a computer system included in a grid computer network. The configuration element entry includes data used by a program included in the computer system in the grid computer network to locate and access application specific customization data associated with an application on the computer system. The system and method also includes accessing the entries in the definition scheme for the application such that the program accesses the customization data using the definition scheme and the configuration element entries.

Abstract: A method includes receiving, from an application process, a document specifying a communication protocol and a communication channel, reading the document, and instantiating the communications channel with the application process using the communication protocol. The method also includes sending communications to the application process using the communication protocol through the communication channel, and receiving communications from the application process using the communication protocol through the communication channel.

Abstract: A method and a graphical user interface (GUI) presents a set of services managing a portion of a computer grid, the GUI including a matrix-like structure with columns and rows, each column representing a computer from a set of computers in the computer grid, each computer from the set of computers having a grid manager, and each row representing a grid manager or other application service, positions of labels in the structure indicating which computer currently runs which grid manager or other application service. The GUI also includes a column representing a first computer from the set of computers running a first grid manager, and one or more columns representing one or more computers from the set of computers running one or more grid managers having an inferior relation with the first grid manager.

Abstract: A graphical user interface (GUI) includes a graph with vectors and nodes for visualizing a computer grid, the nodes representing computers running grid managers and the vectors representing relations between pairs of grid managers, each of the relations defining a first grid manager to be superior to a second grid manager and a vector points from a node representing the first grid manager to a node representing the second grid manager. The GUI also includes, for each node, an expandable structure showing computer grid applications running on a computer represented by the node, and an event handler configured to receive a request to view management services running on a computer and to generate a display showing the management services running on the computer.

Abstract: A method includes requesting a computer in a network to install a first application and execute the first application, transferring one or more files to the computer, making modifications to the computer to install and execute the first application on the computer while recording the modifications, the modifications comprising allocating a resource to the first application, halting the first application on the computer, and reversing a portion of the modifications to the computer according to the recorded modifications. The reversing includes deallocating the resource from the first application. The method also includes making modifications to the computer to install and execute a second application on the available computer while recording the modifications, where the modifications include allocating the resource to the second application.

Abstract: A network of grid managers includes a first computer linked to a second computer, the first computer having a first grid manager and the second computer having a second grid manager, the first and second grid managers handling at least locating, reserving, allocating, monitoring, and deallocating one or more computational resources for an application, the grid manager, upon receipt of a command, loads new instructions to modify current instructions residing in the service that modifies a behavior of the service without restarting the service.