Abstract: The present disclosure provides a method of scheduling data processing at a pipelined data processing engine, and a command scheduler for scheduling data processing at the pipelined data processing engine. The command scheduler determines whether a first data stream is locked to the pipelined data processing engine based on a status of a current data frame of the first data stream in the pipelined data processing engine. The command scheduler will schedule a next data frame of the first data stream to the data processing engine if the first data stream is not locked to the pipelined data processing engine, or it will postpone the scheduling of the next data frame of the first data stream if the first data stream is locked to the pipelined data processing engine.

Abstract: Methods, computer program products, and systems are presented. The methods include, for instance: synchronizing, by one or more processor, first virtual machine(s) with a second virtual machine, the synchronizing comprising mirroring first input events to the first virtual machine(s); and customizing, by the one or more processor, the first virtual machine(s), the customizing comprising suspending the synchronizing and facilitating sending of second input events to the first virtual machine(s). In one embodiment, the synchronizing comprises cloning the second virtual machine to create the first virtual machine(s), the cloning comprising initializing the first virtual machine(s), and copying second virtual machine state information to the first virtual machine(s).

Abstract: A method of executing a job in a computer system having a plurality of different allowed configurations comprises: receiving the job; identifying one or more representative samples of the job; executing the one or more representative samples in an experimental phase, the or each representative sample being executed with each different allowed configuration of the computer cluster; collecting execution data including speed of execution and energy used in each configuration during execution of the one or more representative samples and selecting a suitable configuration for the job taking speed of execution and energy use into account; and executing the remainder of the job in a completion phase with the suitable configuration.

Abstract: Multi-threaded processing with reduced context switching is disclosed. Context switches may be avoided through the use of pre-emption notification, a pre-emption wait time attribute and a no-context-save yield.

Abstract: In one embodiment, a method includes empirically analyzing a set of active reservations and a current set of consumable resources belonging to a class of consumable resources. Each active reservation is of a managed task type and includes a group of one or more tasks requiring access to a consumable resource of the class. The method further includes, based on the empirically analyzing, clocking the set of active reservations each clocking cycle. In addition, the method includes, responsive to the clocking, sorting a priority queue of the set of active reservations.

Abstract: An apparatus and method are provided for allocating resources to a plurality of threads to perform a service. In use, a request for service is received. At least one of a plurality of resources is allocated to the threads. Further, the service is performed with the threads, utilizing the allocated at least one resource.

Abstract: A method, system and computer program product for emulation of a virtual application bookmark on a host desktop. When a user launches a VM (running Windows™), shortcuts to user applications are created on the desktop. In Parallels™ Desktop™, the VM shortcuts and MAC bookmarks can both be seen and be functional. A special file is created in MAC OS X with Windows™ icon. The file is associated with the VM from which the file was created, and resource link corresponding to the original shortcut. The bookmark thus contains a reference to an application icon in the VM, which the user can launch from the host OS by clicking on it.

Abstract: Configurable per-task state counters for processing cores in multi-tasking processing systems are disclosed along with related methods. In part, the disclosed embodiments include a work scheduler and a plurality of processing cores. The work scheduler assigns tasks to the processing cores, and the processing cores concurrently process multiple assigned tasks using a plurality of processing states. Further, task state counters are provided for each assigned task, and these task state counters are incremented for each cycle that the task stays within selected processing states to generate per-task state count values for the assigned tasks. These per-task state count values are reported back to the work scheduler when processing for the task ends. The work scheduler can then use one or more of the per-task state count values to adjust how new tasks are assigned to the processing cores.

Abstract: A method for tuning workflow settings in a distributed computing workflow comprising sequential interdependent jobs includes pairing a terminal stage of a first job and a leading stage of a second, sequential job to form an optimization pair, in which data segments output by the terminal stage of the first job comprises data input for the leading stage of the second job. The performance of the optimization pair is tuned by determining, with a computational processor, an estimated minimum execution time for the optimization pair and increasing the minimum execution time to generate an increased execution time. The method further includes calculating a minimum number of data segments that still permit execution of the optimization pair within the increased execution time.

Abstract: The disclosed embodiments provide a system that facilitates the development and execution of a software program. During operation, the system provides a mechanism for restricting a variable to a runtime context in the software program. Next, the system identifies the runtime context during execution of the software program. Finally, the system uses the mechanism to prevent incorrect execution of the software program by ensuring that a closure capturing the variable executes within the identified runtime context.

Abstract: Described herein are systems for monitoring memory allocation and deallocation associated with compact virtual machine (CVM) instances operating on a host environment. In some cases, the systems output a tree structure that is viewable by a user to understand the native memory usage of the CVM instance stemming from java frames execution. In some particular instances, the tree structure may present to the user a code line by code line analysis of the memory allocation and deallocation.

Abstract: Provided are techniques for multiple stage workload management. A staging queue and a run queue are provided. A workload is received. In response to determining that application resources are not available and that the workload has not been previously semi-started, the workload is added to the staging queue. In response to determining that the application resources are not available and that the workload has been semi-started, and, in response to determining that run resources are available, the workload is started. In response to determining that the application resources are not available and that the workload has been semi-started, and, in response to determining that the run resources are not available, adding the workload to the run queue.

Abstract: Hypercalls for a host machine and guest machines to a hypervisor are intercepted and routed to the hypervisor for execution on a hardware platform, responsive to the hypercall passing hypercall access rules. The intercepting may be performed by a hypercall access control layer that reloads a hypercall table that maps hypercalls from a host machine and guest machines to hypercall handlers, to identify a hypercall access control handler as the hypercall handler for the host machine and the guest machines. Upon passing the hypercall access rules, the hypercall is passed to the hypervisor.

Abstract: Embodiments of adjustment of a task execution plan at runtime by a task execution engine configured to receive a plan compilation task, the plan compilation task comprising a task execution plan, are provided. An aspect includes receiving a first plan compilation task by the task execution engine through a plan compilation interface. Another aspect includes modifying a task execution plan of the first plan compilation task in response to receiving a second plan compilation task by the task execution engine, the second plan compilation task comprising a task execution plan for modifying the task execution plan of the first plan compilation task. Yet another aspect includes reading a next task in the task execution plan of the first plan compilation task and initiating the next task by the task execution engine.

Abstract: A host machine may host a virtual machine. Virtual machine reboot information, used to reboot the virtual machine in the event of a failure or restart of the virtual machine, may be identified (e.g., file system metadata buffers, a virtual non-volatile random access memory log, user data buffers, and/or data used to reboot the virtual machine such as to perform a reboot mounting operation and/or a reboot replay operation of a volume of data associated with the virtual machine). The virtual machine reboot information may be cached within relatively fast host memory of the host machine (e.g., instead of merely within a relatively slower hard drive or other storage device). In this way, the cached virtual machine reboot information may be quickly retrieved so that the virtual machine may be rebooted in a relatively shorter amount of time.

Abstract: A method and system for deploying a component of a composite application in a data center. An example method includes receiving access data from a traffic manager implementing global load balancing, where the access data is related to client accesses to a composite web application including multiple components hosted by virtual machines in a virtualization infrastructure. The method further includes determining, based on the access data, determining that one of the components of the composite web application receives client accesses that exceed a threshold, identifying a new geographic location for the client accesses that exceed the threshold, instructing the virtualization infrastructure to deploy a new virtual machine at the new geographic location, the deployed virtual machine implementing the determined component of the composite web application, and providing information pertaining to the deployed virtual machine to the traffic manager.

Abstract: Systems, methods, and software are described herein for operating a data management system, including a virtual machine agent running within a virtual machine responding to an attach-triggering event, determining selected storage volumes to be attached to the virtual machine based on a request generated by the virtual agent in response to the attach-triggering event, and dynamically attaching the selected storage volumes to the virtual machine.

Abstract: A streams manager monitors performance of parallel portions of a streaming application implemented in multiple virtual machines (VMs). When the performance provided by the multiple VMs is no longer needed, one or more of the VMs can be torn down. The performance of the VMs is monitored. When the least performing VM can be torn down, it is torn down. When the least performing VM cannot be torn down, information regarding a better performing VM is gathered, and it is determined whether the least performing VM can be made more similar to the better performing VM. When the least performing VM can be made more similar to the better performing VM, the least performing VM is changed to improve its performance, and the better performing VM is torn down.

Abstract: A streams manager monitors performance of parallel portions of a streaming application implemented in multiple virtual machines (VMs). When the performance provided by the multiple VMs is no longer needed, one or more of the VMs can be torn down. The performance of the VMs is monitored. When the least performing VM can be torn down, it is torn down. When the least performing VM cannot be torn down, information regarding a better performing VM is gathered, and it is determined whether the least performing VM can be made more similar to the better performing VM. When the least performing VM can be made more similar to the better performing VM, the least performing VM is changed to improve its performance, and the better performing VM is torn down.

Abstract: Various embodiments provide an ability to describe, independent of a programming language, one or more interfaces associated with an operating system. Alternately or additionally, a compiler associated with a specific programming language can be configured to map the independent interface description(s) to the specific programming language. In some embodiments, an application can be configured to programmatically determine one or more interfaces of the operating system.