Abstract: Launch configurations of a hardware acceleration device are determined, which minimize hardware thread management overhead in running a program code. Based on received hardware behaviors, the architectural features, the thread resources and the constraints associated with the hardware acceleration device, possible launch configurations and impossible launch configurations are generated. A ranking of at least some of the possible launch configurations may be generated and output, based on how well each of said at least some of the possible launch configurations satisfies at least some of the constraints. Parametric values of said at least some of the possible launch configurations, an explanation why the impossible launch configurations have been determined as being impossible, and one or more strategies for scheduling, latencies and efficiencies associated with the hardware acceleration device, are output.

Abstract: The data processing times of data processing nodes are heterogeneous, and hence the execution time of a whole system is not optimized. A task is executed using a plurality of optimal computing devices by distributing a data amount of data to be processed with a processing command of the task for the plurality of optimal computing devices depending on a difference in computing power between the plurality of optimal computing devices, to thereby execute the task in a distributed manner using the plurality of optimal computing devices.

Abstract: Improving execution for a served application by identifying a network resource having a partial portion of the application's necessary computing resource requirements, deploying the served application to the network resource, monitoring the execution of the application and storing a checkpoint state of the served application prior to exhausting the resources. Further, by identifying a network resource having a greater portion of the computing resources; and deploying the checkpoint state of the served application to that network resource.

Abstract: A device may obtain ticket data relating to a set of tickets, and process the ticket data to generate a ticket analysis model that is a clustering based natural language analysis model of natural language text associated with tickets of the set of tickets. The device may classify the set of tickets using the ticket analysis model, may determine an automation plan for at least one class of ticket determined based on classifying the set of tickets, and may implement the automation plan to configure an automatic ticket resolution or ticket generation mitigation procedure for the at least one class of ticket. The device may receive a ticket after configuring the automatic ticket resolution or ticket generation mitigation procedure, may classify, using the ticket analysis model, the ticket into the at least one class of ticket, and may automatically implement a response action for the ticket based on classifying the ticket and using the automatic ticket resolution or ticket generation mitigation procedure.

Abstract: Embodiments for providing compare and swap (CAS) functionality to key value storage to allow multi-threaded applications to share storage devices and synchronize multiple concurrent threads or processes. A key-value application programming interface (API) is modified to include a CAS API in addition to the standard Put and Get APIs. The CAS function uses a key, expected old value, and new value to compare and swap an existing key value only if its current value equals the expected old value. Hash values of the key value and expected old value may be used by the CAS function to improve performance and reduce bandwidth.

Abstract: Systems and methods are provided for seamless and automatic upgrade of access points in a facility with minimal disruption, particularly, for important users of a network. The access points may be ranked, and the rankings may be aggregated to identify a sub-region of the facility with the lowest ranked access points. Various sub-sets of the access points in the identified sub-region can be updated separately so that spatial coverage by the other sub-sets maintains continuous access within the sub-region during the updates. Updates to later sub-sets and/or sub-regions can be performed contingent on the success of the earlier updates. In this way, updates can be verified on access points used by lower-priority users such as guest users of the network to avoid disruption of access to more important users such as executives or safety workers on the network.

Abstract: There is described a method for Java Card application memory footprint optimization which relies on the separation in advance of the code related to the personalization from the rest of the code. It allows this code to perform the personalization of an application installed from a main package while being itself included and installed from a separated package dedicated to the personalization, namely the Card Personalization Specifications (CPS) package. This way, the CPS package and all the code inherent to the personalization can be removed once all the personalization steps have been completed.

Abstract: A computer-implemented method, apparatus, and computer program product for scheduling a process to run on a processor. A determination is made whether a process running on a processor is making an input/output operation system call. The process is de-scheduled from running on the processor in response to a determination that the process is making an input/output operation system call. The process is allowed to continue running on the processor in response to a determination that the process is not making an input/output system call.

Abstract: Systems, methods, and software described herein facilitate servicing of data requests based on quality of service assigned to processing jobs. In one example, a method of prioritizing data requests in a computing system based on quality of service includes identifying a plurality of data requests from a plurality of processing jobs. The method further includes prioritizing the plurality of data requests based on a quality of service assessed to each of the plurality of processing jobs, and assigning cache memory in the computing system to each of the plurality of data requests based on the prioritization.

Abstract: An apparatus (2) with multiple processing elements (4, 6, 8) has shared transactional processing resources (10, 50, 75) for supporting processing of transactions, which comprise operations performed speculatively following a transaction start event whose results are committed following a transaction end event. The transactional processing resources may have a significant overhead and sharing these between the processing elements helps reduce energy consumption and circuit area.

Abstract: In one implementation, a method of sharing a physical device between multiple virtual machines is provided. The method includes receiving, from a first virtual machine, a request to access a physical device of a computing device. The method also includes determining, by a processing device, whether the physical device of the computing device is available in view of power state information associated with the physical device of the computing device. The method further includes in response to determining that the physical device of the computing device is available, assigning the physical device to the first virtual machine.

Abstract: One embodiment provides for a computer-implemented method comprising instantiating a synchronization primitive to control access to a resource, acquiring the synchronization primitive at a first thread, the first thread having a first priority, associating a turnstile with the synchronization primitive, setting an inheritor of the turnstile to the first thread, attempting to acquire the synchronization primitive at a second thread while the synchronization primitive is held by the first thread, the second thread having a second priority, adding the second thread to a wait queue of the turnstile; and in response to determining that the second priority is higher than the first priority, increasing the priority of the first thread to the second priority.

Abstract: Combining specifications of dataflow graphs includes receiving: a first dataflow graph specification that specifies two or more components connected by links representing flows of data, and a second dataflow graph specification that specifies at least one component, and at least one sub-graph interface. The sub-graph interface includes at least one flow junction representing a connection between: (1) a flow of data outside the sub-graph interface, and (2) a flow of data inside the sub-graph interface. The method includes processing information including the first dataflow graph specification and the second dataflow graph specification, to generate a combined dataflow graph specification, including: identifying an association between the sub-graph interface and the first dataflow graph specification, for at least a first flow junction, determining a direction associated with transferring a value of a descriptor, and transferring a value of a descriptor according to the determined direction.

Abstract: A pneumatically actuated computer input device comprises an input device circuit member having a hollow body, at least one switch member within the hollow body adapted to send electronic signals to a computer terminal, and a pneumatic actuator within the hollow body adapted to be inflated and press against and activate the at least one switch member. A pneumatic hose connected at a distal end to the pneumatic actuator and extending outward from the hollow body at a predetermined length and adapted to supply air thereto, and a foot pedal connected to the pneumatic hose at an opposite end from the distal end and adapted to control the air supply through the pneumatic hose and to the pneumatic actuator, such that when the foot pedal is pressed the pneumatic actuator is filled with air, expands, and presses against the at least one switch member which then sends an electronic signal to the computer terminal.

Abstract: A method includes obtaining feature generation code from, which is configured to determine features relating to input data. The method further includes obtaining data grouping code, which is configured to generate training data by determining a plurality of data groupings for the features relating to the input data. The method further includes obtaining modeling code, which is derived at least in part by applying one or more machine learning algorithms to the training data. The method further includes applying a model wrapper code to the feature generation code, the data grouping code, and the modeling code to generate a model wrapper and deploying the model wrapper such that the model wrapper may receive a first application programming interface (API) call including an input data value, determine a score relating to the input data value, and send a second API call including the score in response to the first API call.

Abstract: A device may connect to a physical device. A device may obtain a device description document relating to the physical device based on connecting to the physical device. The device may parse the device description document to identify one or more components of the physical device. The device may select, from a repository of generic textual model components, one or more generic textual model components corresponding to the physical device based on parsing the device description. The device may generate, based on selecting the one or more generic textual model components, a device driver model element that includes the one or more generic textual model components. The device driver model element may be used to control the physical device or receive information from the physical device. The device may provide, via a user interface, information identifying the device driver model element.

Abstract: Technologies for secure I/O include a compute device, which further includes a processor, a memory, a trusted execution environment (TEE), one or more input/output (I/O) devices, and an I/O subsystem. The I/O subsystem includes a device memory access table (DMAT) programmed by the TEE to establish bindings between the TEE and one or more I/O devices that the TEE trusts and a memory ownership table (MOT) programmed by the TEE when a memory page is allocated to the TEE.

Abstract: Systems, apparatuses and methods may provide for technology that assigns a plurality of data portions associated with a workload to a plurality of cores, wherein each data portion from the plurality of data portions is only modifiable by a respective one of the plurality of cores. The technology may further pass a message between the plurality of cores to modify one or more of the data portions in response to an identification that the one or more of the data portions are unmodifiable by one or more of the plurality of cores.

Abstract: A simultaneous multi-threading (SMT) processor core capable of thread-based biasing with respect to execution resources. The SMT processor includes priority controller circuitry to determine a thread priority value for each of a plurality of threads to be executed by the SMT processor core and to generate a priority vector comprising the thread priority value of each of the plurality of threads. The SMT processor further includes thread selector circuitry to make execution cycle assignments of a pipeline by assigning to each of the plurality of threads a portion of the pipeline's execution cycles based on each thread's priority value in the priority vector. The thread selector circuitry is further to select, from the plurality of threads, tasks to be processed by the pipeline based on the execution cycle assignments.

Abstract: A method includes obtaining feature generation code from, which is configured to determine features relating to input data. The method further includes obtaining data grouping code, which is configured to generate training data by determining a plurality of data groupings for the features relating to the input data. The method further includes obtaining modeling code, which is derived at least in part by applying one or more machine learning algorithms to the training data. The method further includes applying a model wrapper code to the feature generation code, the data grouping code, and the modeling code to generate a model wrapper and deploying the model wrapper such that the model wrapper may receive a first application programming interface (API) call including an input data value, determine a score relating to the input data value, and send a second API call including the score in response to the first API call.