Abstract: According to one aspect of the present disclosure, a method and technique for job distribution within a grid environment is disclosed. The method includes: receiving jobs at a submission cluster for distribution of the jobs to at least one of a plurality of execution clusters, each execution cluster comprising one or more execution hosts; determining resource capacity corresponding to each execution cluster; determining resource requirements for the jobs; dynamically determining a pending job queue length for each execution cluster based on the resource capacity of the respective execution clusters and the resource requirements of the jobs; and forwarding jobs to the respective execution clusters according the determined pending job queue length for the respective execution cluster.

Abstract: According to one aspect of the present disclosure, a method and technique for job distribution within a grid environment is disclosed. The method includes: receiving jobs at a submission cluster for distribution of the jobs to at least one of a plurality of execution clusters, each execution cluster comprising one or more execution hosts; determining resource attributes corresponding to each execution host of the execution clusters; grouping, for each execution cluster, execution hosts based on the resource attributes of the respective execution hosts; defining, for each grouping of execution hosts, a mega-host for the respective execution cluster, the mega-host for a respective execution cluster defining resource attributes based on the resource attributes of the respective grouped execution hosts; determining resource requirements for the jobs; and identifying candidate mega-hosts for the jobs based on the resource attributes of the respective mega-hosts and the resource requirements of the jobs.

Abstract: According to one aspect of the present disclosure, a method and technique for data processing in a distributed computing system having a service-oriented architecture is disclosed. The method includes: receiving, by a workload input interface, workloads associated with an application from one or more clients for execution on the distributed computing system; identifying, by a resource management interface, available service hosts or service instances for computing the workloads received from the one or more clients; responsive to receiving an allocation request for the one or more hosts or service instances by the workload input interface, providing, by the resource management interface, address information of one or more workload output interfaces; and sending, by the one or more workload output interfaces, workloads received from the workload input interface to the one or more service instances.

Abstract: A service controller for providing centralized service management, control and placement, as well as guaranteed failover, in a dynamic and distributed computing environment. The service controller works in conjunction with a computing resource supply manager, which supplies the required resources so that the service controller can run the service on the allocated resources. The number of active service instances may dynamically increase or decrease in accordance with changing availability of or demand for resources. The service controller monitors the service instances and takes appropriate action in the event of service instance failures. The service controller is provided with a service director component, which records and tracks, for each service instance which has been started on a particular resource/host, the IP address of such host, through a name location service, thus allowing end-users to locate the service regardless of where the service instances are physically running.

Abstract: According to one aspect of the present disclosure, a method and technique for facilitating the exchange of information between interconnected computing entities is disclosed. The method includes: receiving from a client, by a workload manager, a workload unit of data in need of processing by the client; initiating by the workload manager a persistent storage of the workload unit of data received from the client; without waiting for the initiated storage of the workload unit of data to complete, sending by the workload manager the workload unit of data to a plurality of compute nodes; and responsive to receiving a result of a processing of the workload unit of data by one of the plurality compute nodes, canceling processing by the workload manager of the workload unit of data by a remainder of the plurality of compute nodes.

Abstract: Disclosed herein are systems and methods for a distributed computing system having a service-oriented architecture. The system is configured to receive workloads from client applications and to execute workloads on service hosts. The distributed computing system dynamically assigns the workloads to the applications running on the service hosts, with the workloads being assigned according to the service needs and the availability of service hosts and other resources on the system. The presently disclosed systems and methods provide for high-throughput communications through an asynchronous binary or a synchronous binary communications protocol. Further disclosed embodiments include flexible failover and upgrade techniques, isolation between execution users of the system, virtualization through mobility and the ability to grow and shrink assigned resources, and for a software development kit adapted for the present architecture.

Abstract: Disclosed herein are embodiments of a method and system for facilitating the exchange of information between interconnected processors in environments requiring high performance and high reliability. In an exemplary embodiment, the source sends input to the target and expects output from the target in return. A manager in communication with both the source and the target receives and initiates a storage of the information in nonvolatile memory. The manager concurrently forwards the information to its proper destination. If the manager receives output from the target before completion of the input storage, the manager cancels the input storage because it is no longer needed to ensure system reliability. If the manager receives acknowledgement from the source that the target output has been received before completion of the output storage, the manager cancels the output storage because it is no longer needed to ensure system reliability. Related embodiments are also described.

Abstract: Systems and methods for scheduling jobs in a networked computing grid having job service components for creating jobs having certain resource requirements are described. One embodiment includes resource manager components communicatively linked with the job service components, each resource manager component being configured to manage a plurality of resources. Such a system further includes information components communicatively linked with the job service components and the resource manager components, the information components being configured to aggregate resource data obtained from the resource manager components. The jobs include a scheduling mechanism for querying the information components for resource data, comparing the resource data against the resource requirements and independently scheduling the job to the resource manager components. A resource reservation is also provided for reserving resources for the jobs.

Abstract: A method for scheduling jobs in a networked computing grid is described. The method includes scheduling jobs to meet goals related to execution of the jobs within the computing grid. The jobs may be scheduled by evaluating job execution goals against available resource slots provided by the computing grid. Also disclosed are related methods for scheduling jobs in which the jobs are submitted to service classes having defined job execution goals, and wherein the jobs are scheduled according to the job execution goals of the service classes.

Abstract: Presented herein are systems and methods for checking the integrity of data transmissions between or within one or more digital processing systems by identifying a data characteristic that is likely to change if there is an error in transmission. According to one embodiment, data messages are modified to achieve a selected characteristic according to a predetermined protocol, and changes to the data are recorded in a longitudinal check code (LCC) word, which is used by the receiver to decode the data message and restore the original data.

Abstract: Presented herein are systems, methods, and devices for transmitting data comprising by comparing a new message with a buffered message representative of a previous message, recording an integer number of differences between the new message and the buffered message, generating a diff-word representative of the integer number of differences between the new message and the buffered message, appending the diff-word to the new message, and sending the new message with the diff-word to a receiver. The diff-word is used by the receiver to determine whether a received new message has the expected number of differences, and to determine whether to accept the new message.

Abstract: A system and method for scheduling jobs in a multiprocessor machine is disclosed. The status of resources, including CPUs on node boards and associated shared memory in the multiprocessor machine is periodically determined. The status can indicate the resources available to execute jobs. This information is accumulated by the topology-monitoring unit and provided to the topology library. The topology library also receives a candidate host list from the scheduling unit which lists all of the resources available to execute the job being scheduled based on non-trivial scheduling. The topology library unit then uses this to generate a free map F indicative of the interconnection of the resources available to execute the job. The topology monitoring unit then matches the jobs to the resources available to execute the jobs, based on resource requirements including shape requirements indicative of interconnections of resources required to execute the job.

Abstract: Presented herein are systems and methods for checking the integrity of data transmissions between or within one or more digital processing systems by identifying a data characteristic that is likely to change if there is an error in transmission. According to one embodiment, data messages are modified to achieve a selected characteristic according to a predetermined protocol, and changes to the data are recorded in a longitudinal check code (LCC) word, which is used by the receiver to decode the data message and restore the original data.

Abstract: Systems and methods for scheduling jobs in a multiprocessor machine are disclosed. The status of resources in the multiprocessor machine is periodically determined. The status indicates the resources available to execute jobs. This information is accumulated by the topology-monitoring unit and provided to the topology library. The topology library also receives a candidate host list which lists all resources available to execute the job being scheduled based on non-trivial scheduling. The topology library unit generates a free map F indicating the interconnection of the resources available to execute the job. The topology monitoring unit matches jobs to the resources available to execute the jobs, based on resource requirements including shape requirements indicative of interconnections of resources required to execute the job. The topology monitoring unit dispatches the job to the portion of the free map F which matches the shape requirements of the job.

Abstract: Disclosed herein are systems and methods for a distributed computing system having a service-oriented architecture. The system is configured to receive workloads from client applications and to execute workloads on service hosts. The distributed computing system dynamically assigns the workloads to the applications running on the service hosts, with the workloads being assigned according to the service needs and the availability of service hosts and other resources on the system. The presently disclosed systems and methods provide for high-throughput communications through an asynchronous binary or a synchronous binary communications protocol. Further disclosed embodiments include flexible failover and upgrade techniques, isolation between execution users of the system, virtualization through mobility and the ability to grow and shrink assigned resources, and for a software development kit adapted for the present architecture.

Abstract: Disclosed herein are embodiments of a method and system for optimizing resource management in a distributed computing environment through the use of a resource conductor. An application managed by an application manager requires resources managed by a resource manager. A resource conductor in communication with both the application manager and the resource manager receives from the application manager a processing specification for the application and workload associated with the application. The processing specification provides the resource conductor with information needed to determine the type and quantity of resources appropriate for processing the workload associated with the application. The resource conductor adjusts the quantity of resources allocated to the application by communicating with the resource manager.

Abstract: Disclosed herein are embodiments of a method and system for facilitating the exchange of information between interconnected processorsin environments requiring high performance and high reliability. In an exemplary embodiment, the source sends input to the target and expects output from the target in return. A manager in communication with both the source and the target receives and initiates a storage of the information in nonvolatile memory. The manager concurrently forwards the information to its proper destination. If the manager receives output from the target before completion of the input storage, the manager cancels the input storage because it is no longer needed to ensure system reliability. If the manager receives acknowledgement from the source that the target output has been received before completion of the output storage, the manager cancels the output storage because it is no longer needed to ensure system reliability. Related embodiments are also described.

Abstract: An application manager for enabling multiple applications to share resources in a shared and distributed computing environment. The disclosed system provides for the specification, representation and automatic analysis of resource requirements of applications in a shared and distributed computing environment. The application manager is provided with service specifications for each application, which defines the resource requirements necessary or preferred to run said application (or more precisely, its constituent application components). In addition, the resources may be required to have certain characteristics and constraints may be placed on the required resources. The application manager works in conjunction with a resource supply manager and requests the required resources be supplied for the application. If there are appropriate and sufficient available resources to meet the particular resource requirements, then the resources are allocated, and the application components mapped thereon.

Abstract: A service controller for providing centralized service management, control and placement, as well as guaranteed failover, in a dynamic and distributed computing environment. The service controller works in conjunction with a computing resource supply manager, which supplies the required resources so that the service controller can run the service on the allocated resources. The number of active service instances may dynamically increase or decrease in accordance with changing availability of or demand for resources. The service controller monitors the service instances and takes appropriate action in the event of service instance failures. The service controller is provided with a service director component, which records and tracks, for each service instance which has been started on a particular resource/host, the IP address of such host, through a name location service, thus allowing end-users to locate the service regardless of where the service instances are physically running.

Abstract: A resource manager for managing the sharing of resources among multiple workloads in a distributed computing environment. The resource manager comprises an allocation service component which applies a resource allocation policy to determine whether the requested resources may be allocated. The resource manager also comprises a dynamic provisioning service component which applies a reprovisioning policy to modify the make-up of the resources in response to unsatisfied resource requests. The allocation of resources to consumers may be determined in accordance with the resource ownership and share rights of the consumer on behalf of which the application or workload is being demanded.