281results about "Multiprogramming arrangements" patented technology

A system may include a robotic control circuit configured to control one or more robotic functions of a robot. A system may include a plurality of sensors configured to collect data. A system may include a governance analysis circuit configured to analyze the data and select one or more governance frameworks based on the analyzed data. A system may include a governance model circuit configured to generate a model that applies the one or more governance frameworks to determine one or more governance actions, wherein the robotic control circuit is configured to control the one or more robotic functions in accordance with the one or more governance actions, wherein the robotic control circuit, the governance analysis circuit, and the governance model circuit are integrated on a single substrate.

The invention particularly relates to a distributed time series data cross-node atomic writing and transaction management method. According to the distributed time series data cross-node atomic writing and transaction management method, a coordination node receives a time series writing request, generates a unique TID and issues the unique TID; the storage node writes the data into a pre-write log WAL, marks the data as pending, and establishes mapping between the TID and log entries; the cooperative node sends a pre-submission instruction, and the node responds after being ready; and if verification is passed, broadcasting submission is carried out, a formal storage update version is written, otherwise, rollback is carried out, operation is cancelled according to WAL, and finally associated logs and states are cleared. According to the distributed time series data cross-node atomic write-in and transaction management method, end-to-end atomicity, consistency, low delay and rapid recovery of cross-node time series data write-in are achieved, a write-in path only needs one-time WAL write-in, synchronous blocking is avoided, and the concurrent throughput performance is remarkably improved; and the scheme does not invade an upper-layer application interface, and can be reused in any distributed storage system supporting the WAL.

The invention relates to a register management method and device, electronic equipment and a storage medium. The method comprises the following steps: allocating a shared on-chip storage space for registers of a plurality of processing units in processing equipment; setting a write-in mark corresponding to the register of each processing unit in an array of the processing equipment; in response to execution of the first command, reading register data of the target register from the off-chip storage space, storing the register data in the on-chip storage space, and setting a write-in mark corresponding to the target register in the array to be valid; and in response to execution of the second command, determining an effective target write-in mark from the array, reading register data corresponding to the target write-in mark from the on-chip storage space, and writing the register data into a target register of the processing unit indicated by the target write-in mark. The embodiment of the invention can improve the operation efficiency of the processing equipment.

The invention relates to the technical field of cloud computing, and discloses a multi-cloud certificate deployment intelligent arrangement engine system, which comprises a unified abstraction layer used for providing a standardized certificate deployment interface; the intelligent arrangement engine is used for analyzing a deployment task dependency relationship and generating an optimal deployment path; the distributed transaction management module is used for coordinating deployment transactions of the cross-cloud platform and ensuring the consistency of operation; and the strategy optimization and learning module is used for optimizing the deployment strategy based on the historical deployment data. According to the method, the cloud platform difference is shielded through the unified abstraction layer, the consistent deployment experience is provided, and the adaptation workload in the multi-cloud environment can be remarkably reduced; the deployment path is optimized through the intelligent arrangement engine, so that the deployment time can be remarkably shortened; through distributed transaction management, transaction consistency of cross-cloud deployment is ensured, and strong fault recovery capability is provided. Through multi-objective optimization and online learning, the system can automatically balance the cost, the performance and the reliability, so that the resource utilization rate is greatly improved.

The embodiment of the invention provides a session processing method and device based on a large model, electronic equipment and a storage medium, and the method comprises the steps: obtaining a problem session instruction inputted by a user, and the problem session instruction at least comprises a target session identifier; according to pre-stored historical storage information, target historical information corresponding to the target session identifier is determined, and the historical storage information at least comprises the session identifier and historical information corresponding to the session identifier; determining input information corresponding to the target session identifier according to the target session identifier and the target historical information; the input information is input into the session processing model to obtain the answer corresponding to the question session instruction, and the session processing model is obtained by training a large model through sample data, so that historical information can be comprehensively considered, more accurate answer information can be provided, errors are reduced, and development efficiency is improved.

The invention provides a code-free SaaS application rapid development method and system based on data source driving, and the method comprises the steps: constructing a forward business process directed acyclic graph, defining compensation association and reality binding for a forward operation node of the forward business process directed acyclic graph, and generating a visual transaction model; analyzing the transaction model to generate a transaction step sequence, constructing a finite state set and a state transfer function based on the sequence, and packaging the finite state set and the state transfer function into an SAGA process state machine definition object; generating a global transaction ID when a transaction life cycle is started, constructing a transaction context object in combination with the state machine definition object, and serializing to generate an initial transaction context instance; and loading and deserializing the instance, starting a state-driven execution loop, terminating by iterating single-step state transition when the current state belongs to a termination state set, judging a final result of the transaction, and archiving an audit log. According to the method, transaction process modeling and formalized state driving execution are fused, and the method is suitable for SaaS application development of cross-heterogeneous data sources.

The invention provides a non-intrusive asynchronous processing method based on transaction event monitoring. An asynchronous event is triggered through a transaction event monitor after a database transaction is submitted, it is ensured that asynchronous operation strictly follows a transaction life cycle, and the risk of data inconsistency caused by dislocation of a thread pool task and a transaction time sequence in a traditional scheme is fundamentally avoided; by intercepting a business method carrying asynchronous execution annotations and automatically extracting parameters to generate an asynchronous request event, physical isolation of business logic and an asynchronous mechanism is realized, and architecture pollution is eliminated; and further dynamically routing to an instant execution or persistent execution module according to the policy type annotated and declared, so as to provide elastic execution capability for different service scenes. Therefore, on the premise of guaranteeing transaction atomicity, a zero-intrusive and extensible asynchronous processing framework is constructed, and the reliability and maintainability of the system are remarkably improved.

To provide a method, a system, and a computer program, for rollback of services with global variable changes.SOLUTION: A method include: detecting that at least two of multiple services in a transaction are executed to change a value of a first global variable; acquiring tracking information to indicate a first order in which the at least two services change the value of the first global variable during execution of the multiple services; and determining, in response to a failure of the transaction, a rollback execution plan for multiple compensation services at least partially based on the tracking information. The multiple compensation services are configured to compensate for the multiple services respectively. The method further includes executing the multiple compensation services according to the rollback execution plan.SELECTED DRAWING: Figure 15

The present application belongs to the technical field of supply chain collaborative management, and discloses a method and system for bi-directional synchronization and modification of logistics and cargo information. Before the supplier completes the delivery, the method allows the supplier to modify or delete the bound logistics information; the operation instruction synchronously triggers the state rollback of the bound cargo information, so that the cargo information returns to the state of waiting for delivery and is unbound; all changes are ensured to be complete and persistent through atomic transactions; finally, based on the transaction success signal, the system pushes the differentiated synchronization information to the related demand side in real time. The system corresponds to the instruction receiving, collaborative processing, transaction management and real-time synchronization modules. The present application solves the problem of data inconsistency and view asynchronization caused by information modification under multi-party collaboration, realizes flexible adjustment of logistics information, automatic recovery of business state and real-time consistency of multi-party view, and significantly improves the reliability and efficiency of supply chain collaboration.

Methods, devices, systems, and manufactured articles for the automatic processing of preliminary peripheral data are disclosed. An exemplary device includes at least one programmable circuit for using a machine learning model to select a first application or a second application based on context information associated with at least one of the first application, the second application, or an input signal from a peripheral device, and to forward the input signal to the selected first application or second application.

A computer system performs a processing operation in real time on an aggregated value stored in a counter of a set of counters corresponding to time periods associated with events. The computer system maintains the set of counters usable to store aggregated values of an event metric. In response to detection of the events, the computer system stores, in a database, event details for the received events and updates those counters of the set of counters that correspond to time periods associated with the received events. In response to receipt, in a current time period, of an update to a particular event of the received events associated with a previous time period, the computer system, in real time, retrieves, from the set of counters, a particular aggregated value of the event metric for the previous time period and performs a processing operation based on the particular aggregated value.

The invention discloses a method and a system for real-time monitoring and automatic exception processing of micro-magnetic calculation. The method comprises the following steps of: acquiring current calculation state information of a main program which is executing a micro-magnetic calculation task; judging whether the current calculation state is abnormal or not according to the acquired current calculation state information; if the current calculation state is not abnormal, executing a current calculation state storage action on the main program which is executing the micro-magnetic calculation task; and if the current calculation state is abnormal, an abnormal state repairing action is automatically executed on the main program executing the micro-magnetic calculation task. The method aims at solving the problem that in micro-magnetic calculation, due to model setting, solver stability and other reasons, human interference must be needed when an abnormal state occurs, then high-flux micro-magnetic intelligent calculation without human supervision and interference is achieved, a magnetic microstructure and magnetic performance database is rapidly established, and rapid research and development of magnetic materials are assisted.

The embodiment of the invention provides a transaction data management method and device, equipment, a storage medium and a program product, and relates to the field of database management. The method comprises the steps of determining a statement feature value of a first SQL statement, wherein the statement feature value comprises at least one of an operation type, an access object identifier and a grammar feature; under the condition that the target transaction is successfully submitted, a target transaction feature identifier corresponding to the target transaction is determined according to the statement feature values of the SQL statements, the transaction corresponding to the SQL statements is the target transaction, and the first SQL statement is any one SQL statement in the SQL statements; and binding the target transaction feature identifier and target transaction statistical data corresponding to the target transaction, and storing the target transaction statistical data in a storage space, wherein the target transaction statistical data comprises at least one of a transaction identifier, submission time and client information. According to the method provided by the invention, the transaction data management capability is improved.

The application discloses a management platform system for supply chain finance and a non-sensing upgrade method thereof, has the effects of local pilot, small flow verification, automatic observation, rapid rollback, and protects the consistency of in-transit instances and the continuity of key transactions. The system comprises: an infrastructure layer, which provides process arrangement, rule and model execution, event driving, configuration management, task compensation, security authentication and full-link observability, supports the non-sensing upgrade core mechanism including multi-version parallel running, shadow verification, gray release and automatic rollback of the upper-layer business, an interaction layer, which provides unified authentication, permission and configurable menu, supports displaying differentiated functions according to organizations / roles, a domain service layer, which is used for encapsulating business logic operations in the platform, coordinating multiple parties to complete cross-entity transactional tasks, and ensuring the consistency and integrity of business rules, and a fund and channel adaptation layer, which abstracts multiple protocols by using a contract / interface definition language, automatically generates basic adapters, and unifies strategies.

A machine learning accelerator (MLA) implemented on a semiconductor die includes a computing mesh of interconnected compute elements. The compute elements execute a program of instructions to implement a machine learning network according to a static schedule for execution of the instructions. The compute elements include gap counters. The number of cycles between any two instructions (i.e., the gap count) in a statically scheduled program is known and fixed. A gap counter counts cycles during execution and must reach the expected gap count before the later instruction can be executed. Synchronization between different processing elements may be maintained by suspending counting for a period of time.

A disclosed event notification management method and service for use with a multi-node, hyper-converged infrastructure (HCl) cluster employs a service running on a corresponding node to monitor one or more parameters associated with event notifications sent to a cluster. The service may adjust a node's notification interval based on the monitored parameters. The parameters may include a last event notification successful (LENS) parameter indicating whether a recent notification was successful and a severity parameter determined based on an urgency type of a plurality of pending notifications within an event queue. The service may employ a first formula to decrease the notification interval following a successful previous notification and a second formula to increase the notification interval following an unsuccessful notification. The magnitude by which the notification interval is adjusted may be influenced by the severity parameter, e.g., to aggressively reduce the notification interval following successful notifications when urgency is high.

Systems for fast views of items in file directories or file folders when interacting with a cloud-based service platform. A server in a cloud-based environment interfaces with one or more storage devices to provide storage of shared content accessible by two or more user devices. A file tree request to view the file directory or file folder of a particular sought after item is issued from an application operating on one of the user devices. Additional file tree items in a file tree hierarchy are prefetched by the cloud-based service platform. The application closes the file tree metadata stream after receiving the portion of the file tree that pertains to the particular item and before receiving the entirety of the metadata pertaining to all of the file tree metadata of all of the items in the directory or folder that contains the particular sought after item.

Scientific instrument support systems and related methods, computing devices, and computer readable media are disclosed herein. For example, in some embodiments, a scientific instrument support apparatus includes a first logic element to generate a first command frame including a flag having a first value; and a second logic element to transmit the first command frame to a first device. The first value indicates that the first device should ignore an operation indicated by the first command frame. The scientific instrument supporting device comprises a third logic element which is used for determining whether a confirmation signal indicating successful operation is received or not; and a fourth logic element for generating a second command frame including a flag having a second value upon receipt of the acknowledgement signal. The second value indicates that the first device should perform an operation indicated by the second command frame.

The application provides a SAGA-based distributed transaction implementation method and device. The method comprises the following steps: binding a SAGA transaction coordinator with a composite service to obtain a bound composite service, the SAGA transaction coordinator is used for coordinating distributed transaction related layer operations of the composite service and plays a role of monitoring transactions; and the distributed transaction is executed by using the bound composite service. The SAGA transaction coordinator is bound with the composite service to obtain the bound composite service and execute the distributed transaction, so that a transaction coordinator does not need to be separately deployed, the problem that, when a large number of transactions occur, the performance of the transaction coordinator server is greatly stressed due to the fact that each transaction needs to interact with the transaction coordinator, and the non-functional guarantee such as high availability and high performance of the transaction coordinator when the transaction coordinator is separately deployed does not need to be considered.

The invention discloses a calculation optimization method and device, a storage medium and a computer program product, and belongs to the technical field of computers. The method comprises the following steps: determining a topological sequence of a computational graph; determining a target operator node in the plurality of operator nodes and a target input node corresponding to the target operator node based on a calculation relationship represented by the topological sequence and the calculation graph, the target operator node being an operator node used for in-situ calculation in the calculation graph, the memory used by the output data of the target operator node is consistent with that used by the input data of the target input node; and executing a calculation task corresponding to the calculation graph based on the target operator node and the target input node. That is, according to the scheme, the nodes capable of performing in-situ calculation in the calculation graph can be automatically identified by the computer equipment. Compared with a method of manually assigning in-situ calculation, the scheme has the advantages that errors are not prone to occurring, efficiency is higher, most of even all nodes capable of conducting in-situ calculation can be found, and therefore the overall calculation performance can be greatly improved.

The present disclosure provides systems, methods, and machine-readable media for concurrent communication. One method includes displaying an interactive application on a first device of a user during a break in an entertainment segment displayed on a second device, sending a request for interaction from the user through the interactive application, pausing the interactive application when the break in the entertainment segment is over, and notifying the user when an additional break begins so the user can commence interacting with the interactive application.

Methods and systems for optimizing neural networks, and in certain embodiments optimizing neural networks for low power edge devices / sensors and providing fault tolerance for neural network execution. Certain disclosed neural networks can change width during inference, such as using one width for one layer, and using a different width for a later layer and such networks may be trained such that the accuracy at any width / depth combination is equivalent to a standard network (that cannot change) of the same width and depth.

The invention discloses a machine learning reasoning call for database query processing. In some embodiments, a method of making a machine learning reasoning call for database query processing may include generating a first batch of machine learning requests based at least on a query to be performed on data stored in a database service, where the query identifies a machine learning service; sending the first batch of machine learning requests to an input buffer of an asynchronous request handler, the asynchronous request handler generating a second batch of machine learning requests based on the first batch of machine learning requests; and obtaining a plurality of machine learning responses from an output buffer of the asynchronous request handler, the machine learning responses being generated by the machine learning service using a machine learning model in response to receiving the second batch of machine learning requests.

The invention provides a transaction processing method and device based on a task chain, which are used for meeting automatic transaction processing requirements before and during a transaction, and fusing a task chain processing mechanism on the basis of automatically deploying an autonomous execution strategy to realize automatic transaction processing, so that a target transaction is deeply analyzed, and the transaction processing efficiency is improved. According to the method, the task chain nodes are divided into different task chain nodes, and the different task chain nodes are embodied, so that manual operation is avoided as far as possible, deep processing elements of the target transaction can be captured, and the target transaction can be executed more efficiently and with high quality through the task contents divided and nested by the different task chain nodes.

1. A processing device comprising: at least one execution unit configured to interleave execution of multiple worker threads, each of the worker threads configured to execute a same set of code to operate on different data sets held in an input buffer of a memory of the processing device, and to output result data to an output buffer, wherein instructions are executed to cause a plurality of operand registers, each associated with one of the worker threads, to input one or more variables that enable each worker to determine where in the input buffer its input data set is located, and where result data should be stored.

To provide a transaction management system capable of maintaining satisfactory reliability of a BC service while managing a resource-sharing status as appropriate among participating organizations in the BC services.SOLUTION: A transaction management system 45000 is a distributed ledger system in which a business network is constituted of multiple nodes 20000. The node 20000 is configured to calculate an upper limit number of transactions that each organization can issue using calculation rules shared among the nodes, and refuse acceptance of the transaction of an organization whose past number of transactions exceeded the upper limit number of transactions among the organizations.SELECTED DRAWING: Figure 1

Systems, methods, apparatuses, and computer program products are disclosed for automatically scaling a deployment based on resource utilization data of a workload executing on the deployment. Resource availability is determined based on the resource utilization data and the deployed current resource allocation. A severity of resource throttling of the workload may be determined based on the resource utilization data, and a scaling factor may be determined based at least on the severity of the resource throttling. At least in response to the resource availability satisfying a predetermined condition having a predetermined threshold, the deployment is scaled based on a scaling factor.