Abstract: Systems and methods for managing rate-controlled and scheduled workloads in distributed computing systems. One example method provides a distributed two-phase algorithm, which utilizes an external timer service and local high-resolution timers to orchestrate time critical workloads on distributed computing systems. The first phase is a planning phase implemented as a recurrent process by means of repeated registrations to the external timer service. During the planning phase, a portion or all of the workload to be performed is broken down into small batches to be assigned to computing instances for execution at a later time. The second phase is a batch execution phase. Each batch planned during the planning phase includes fine-grained instructions as to when each item within the batch is to be executed. The computing instance receiving the batch uses local high-resolution timers to synchronize the execution of items relative to one another.

Abstract: An event processing method provided by an embodiment of the present disclosure determines M events, where each event is associated with a priority and a minimum working time, M is an integer greater than 1, and the minimum working time of each event is less than duration of a time slice; processes a first event within a first time slice, where the first event is an event having a highest priority among the M events; and in the case that processing duration of the first event reaches the minimum working time of the first event, determines to continue to process the first event within the first time slice, or determines to interrupt processing of the first event and processes a second event within the first time slice.

Abstract: In various examples there is a method performed by a controller in Kubernetes cluster. The method comprises: identifying a job to be completed by the cluster, from a plurality of jobs. In response to identifying a job to be completed by the cluster, determining at least one sidecar container associated with the job. In response to identifying a job to be completed by the cluster, determining that the job has been completed by querying a Kubernetes control plane of the cluster. In response to determining that the job has been completed, triggering termination of the sidecar container.

Abstract: Systems and methods include determination of a plurality of event types, determination of a plurality of associated metrics for each of the plurality of event types, generation of synthetic time-series data of each of the metrics associated with the event types for each of the plurality of event type, the synthetic time-series data representing metric anomalies and events of the event type, training of a first system to generate anomaly values based on the synthetic time-series data, training of a second system to classify event types based on the anomaly values, determination of an anomaly value for each data instance of actual time-series data, determination of event types for each determined value, selection of data instances based on the determined values and event types, reception of an event type for each selected data instance, and re-training of the second system based on the selected data instances and event types.

Abstract: A method and device for capability-based split computing is disclosed. The method includes detecting an abnormal status of a user equipment (UE) based on an environment parameter for a device and a performance parameter of the device; transmitting, to a server, a message including UE capability information related to the environment parameter and the performance parameter; receiving, from the server, information related to adjusted content and an adjusted process according to the UE capability information, and executing the adjusted process to receive the adjusted content.

Abstract: A processing apparatus 1 includes a determiner 30 that compares state information, which is information pertaining to states obtained from a field, and performs processing related to a determiner 30 based on information specifying a content of a comparison in the determiner and an operation pertaining to a state transition according to the result of the comparison, and on information pertaining to generation of a state information generator for generating the state information from the information pertaining to the state of the field. The information specifying the contents of the comparison in the determiner 30 and the operation pertaining to the state transition according to the result of the comparison, and the information pertaining to the generation of the state information generator for generating the state information from the information pertaining to the state of the field can be changed regardless of the operation of the determiner 30.

Abstract: In various examples, source archives may be classified, sorted, downloaded, and installed using command files for container image generation. For example, a user may select source archives for inclusion in a container build operation, and the system may classify the source archives as belonging to one of any number of classes and/or sub-classes. Where a size of the source archive is below an associated threshold, the source archive may be included in the container image using native context copy operations. In contrast, where the size of the source archive is above the associated threshold, the source archive may be included in the container image using a local HTTP auto copy operation—e.g., by creating a local HTTP server, and generating a command line such that a container builder retrieves the archive information via the local HTTP server.

Abstract: Disclosed herein is a method for task scheduling for an accelerator pool. The method includes receiving state information of an accelerator in the accelerator pool, generating a list of accelerators capable of being assigned a task based on the state information, calculating the amount of additional power that is to be consumed when the task is assigned to each of the accelerators in the list, and assigning the task to an accelerator in consideration of the calculated amount of additional power to be consumed and conductive heat between accelerators.

Abstract: A computing system includes a DPU subsystem coupled to a host subsystem. The DPU subsystem receives an instruction to provide a service, and determines a host operating load of the host subsystem and a DPU operating load of the DPU subsystem during subsequent time periods. Based on the host operating load and/or the DPU operating load during a first time period, the DPU subsystem configures the DPU subsystem to provide the service, and performs service requests for the service from client system(s). Based on the host operating load and/or the DPU operating load during a second time period that is subsequent to the first time period, the DPU subsystem configures the host subsystem to provide the service in order to migrate the service from the DPU subsystem to the host subsystem, and proxies service requests for the service between the client system(s) and the host subsystem providing the service.

Abstract: An object of the present invention is to provide a SCADA web HMI system that can reduce a load for processing input/output signals in a large-scale system. An HMI server apparatus performs three filtering processing. In first filtering processing, in a case where received block data corresponds to a screen currently displayed on a web browser, a set of input/output signals is extracted from the received block data. In second filtering processing, input/output signals each having a value varied from a previous value are extracted from the set of input/output signals extracted in the first filtering processing. In third filtering processing, input/output signals corresponding to display parts arranged in the screen currently displayed on the web browser, are extracted from the input/output signals extracted in the second filtering processing.

Abstract: Methods and apparatuses for providing a dynamically scalable ADNA Manager with decentralized atomic decision making are described. The decentralized atomic decision making may be performed using atomic deterministic next action (ADNA) task blocks that execute one or more workflow rules and then invoke one or more ADNAs within a pool of ADNAs. The ADNA Manager may identify a first ADNA task block, determine a set of input parameters for the first ADNA task block, detect that a first input parameter of the set of input parameters does not satisfy a qualification rule for the first ADNA task block, identify an exception ADNA task block in response to detection that the first input parameter does not satisfy the qualification rule, store breadcrumb information for the first ADNA task block within a persistence layer prior to the exception ADNA task block being invoked, and invoke the exception ADNA task block.

Abstract: Interactive event streaming can be provided in real-time and selectively transmitted to data consumers based on a rules engine. For example, a computer-implemented method described herein can include receiving a plurality of interaction events from at least one processing application. The method can include modifying the plurality of interaction events to produce a plurality of secured interaction events, wherein personal information of at least one data publisher in each secured interaction event of the plurality of secured interaction events is obscured. The method can further include selecting, based at least in part on a rules engine, at least one particular interaction event from the plurality of secured interaction events. Additionally, the method can include notifying a data consumer of the at least one particular interaction event from the plurality of secured interaction events.

Abstract: Methods, systems, and computer-readable storage media for machine learning database memory use prediction and adaptation. An example method includes determining a sampling interval for an application for sampling memory use by a database for the application. A plurality of historical memory use samples of amounts of memory used by the database are determined for the application based on the sampling interval. The plurality of historical memory use samples are provided for training of a machine learning model to predict memory use for the application by the database for a future time period. A set of current memory use samples are provided to the machine learning model and a memory use prediction for the application for an upcoming time period is received from the machine learning model. A determination is made as to whether to extend memory of the database for the application based on the memory use prediction.

Abstract: A computer-implemented method for planning an allocation of at least one computational task from a computational resource comprised in at least one vehicle to one or more of a plurality of external computational resources in a vehicular communications network. The method comprises obtaining a spatial representation of a region characterising at least one route of a vehicle from a first location to a second location, and data characterising an availability of external computational resources at a plurality of locations in the region, providing at least one computational requirement indication of at least one atomic computational task required by the vehicle during a prospective movement of the vehicle from the first location to the second location, comparing the at least one computational requirement indication to the data characterising the availability of external computational resources at the plurality of locations in the region.

Abstract: Processing in an embodiment can include: receiving a request for a new hardware configuration of a data storage system, wherein the data storage system has a current configuration and is running a workload W1 with a workload pattern P1 having an associated workload periodicity interval, wherein the new hardware configuration is modeled as running an associated workload W2 with the workload pattern P1 and the associated workload periodicity interval, wherein the workload W2 is equal to the workload W1 multiplied by a target scale factor (TSF); in response to the request, determining the new hardware configuration in accordance with the TSF and a plurality of hardware utilizations of a plurality of hardware components in the current configuration while running the workload W1 with the workload pattern P1 having the associated workload periodicity interval; and upgrading the current configuration of the data storage system based on the new hardware configuration.

Abstract: A multi-processor communication framework (MCF) based on Transmission Control Protocol/Internet Protocol (TCP/IP) architecture is defined that can be used for inter-processor communication between heterogenous processing nodes. The MCF includes an application layer, a transport layer, a network layer, and an inter-processor communication (IPC) layer for each processing node. The IPC layer can be used to facilitate data transfer between two processing nodes via a physical channel within an electronic device, or through shared memory. The MCF can support multicast, broadcast, unicast, and zero-copy features like the TCP/IP stack.

Abstract: A memory management method and an electronic device are disclosed. The method includes: starting an application and providing virtual memory to the application; checking a space of the virtual memory occupied while the application is running; checking bitness of a process run by the application when the space of the virtual memory occupied is greater than a first threshold; and performing a process-killing operation on the application based on the bitness of the process run by the application. In the foregoing method, a specific application is checked and killed according to a preset policy by combining multi-dimensional information according to occupancy of virtual memory based on a memory exception occurring in a running application, so as to resolve a memory leak problem.

Abstract: Migrating container-based quantum processes to quantum isolation zones (QIZs) is disclosed herein. In one example, a processor device of a quantum computing device receives a container specification file comprising an indication of a process definition file of a quantum process and an indication of an execution requirement of the quantum process. The processor device determines, based on the execution requirement, that a QIZ provided by the quantum computing device satisfies the execution requirement of the quantum process, wherein the QIZ limits qubit visibility of any quantum process associated with the QIZ to qubits associated with the QIZ. In response to the determining, the processor device allocates one or more qubits of a plurality of qubits associated with the QIZ to the quantum process based on the process definition file, and initiates execution of the quantum process to utilize the one or more qubits, based on the process definition file.

Abstract: Disclosed by the present application are a memory operating-frequency adjustment method, intelligent terminal, and storage medium, comprising: according to operation information of the intelligent terminal, determining memory read/write data of a terminal system; determining the operating mode of the memory according to the memory read/write data; adjusting the operating frequency of the memory according to the operating mode. In the present application, the operating mode of the memory is determined according to the memory read data of the terminal system, and the operating frequency of the memory is adjusted according to the operating mode, improving memory utilization.

Abstract: A method and an apparatus for adjusting an instruction pipeline, a memory and a storage medium. The method for adjusting the instruction pipeline includes: receiving a move-type instruction, wherein the move-type instruction includes a target operand and a first source operand, and indicates to move a data of a source address indicated by the first source operand to a target address indicated by the target operand; receiving an object instruction located after the move-type instruction in an instruction sequence, wherein the object instruction includes an object source operand, indicating that an object operation is performed with a source address indicated by the object source operand; and replacing the object source operand in the object instruction with the first source operand of the move-type instruction to obtain a modified instruction, in response to the target operand of the move-type instruction being identify to the object source operand of the object instruction.