1197results about "Interprogram communication" patented technology

Systems and methods are disclosed comprising techniques for preliminary signal evaluation, such as generating a first session record for a first environment state for the user runtime session, retrieving a second session record for a second environment state for prior user runtime sessions, inputting the first and the second session records into a first generative model to generate a predicted session event set for the user runtime session, selectively determining a prioritized session event from the predicted session event set based on comparing realization parameters of the predicted session event set, inputting the prioritized session event and the first environment state into a second generative model to generate a preliminary session event for execution during the user runtime session, and when the environment state of user runtime session satisfies activation criterions of the preliminary session event, executing the preliminary session event prior to execution of the prioritized session event.

Methods, systems, and products for alert generation and augmentation using a large language model (‘LLM’), including: providing, to the LLM, information describing a monitored deployment; providing, to the LLM, information describing one or more alerts associated with the monitored deployment; and receiving, from the LLM, an enhanced alert, wherein the enhanced alert is based at least in part on at least one of the information describing the monitored deployment and the information describing one or more alerts associated with the monitored deployment.

A data analytics system including an append-only first data store accessible to multiple clients and a second data store is disclosed. The data analytics system can be configurable to, in response to receiving first instructions from a first target system of a first client, the first target system separate from the data analytics system, create a first pipeline between the append-only first data store and the second data store. The first pipeline can be configured according to the first instructions to generate a client-specific data object and store the client-specific data object in the second data store. The data analytics system can be configurable to tear down the first pipeline upon completion of storing the client-specific data object in the second data store.

Techniques are disclosed for a conversational assistant system that employs a machine learning system to act as an intermediary between natural language questions of a user and one or more Application Programming Interfaces (APIs) implemented by a network analysis system. In one example, the conversational assistant system receives API definition files including sets of function definitions and corresponding function descriptions and a natural language question from a user device. The conversational assistant system processes the API definition files and the natural language question to obtain function definitions invoking API endpoints of the APIs. The conversational assistant system sequentially invokes, based on the function definitions, the API endpoints to obtain responses from the APIs implemented by the network analysis system. The conversational assistant system generates, based at least in part on the responses from the APIs, a natural language answer for output at an output device.

This invention provides a universal interface system and its operation method. The universal interface system includes: an upper-layer application algorithm layer (4-1), configured to call functional modules and corresponding interfaces to complete a technical task; a fine-grained interface layer (4-2), configured to provide an abstract and unified fine-grained interface for the upper-layer application algorithm layer and define the input and output standards of each fine-grained interface; a chip adaptation layer (4-3), configured to carry out mapping between the interface specifications of the fine-grained interface and the hardware capabilities of the lower chip execution layer, and to implement the various functions defined by the fine-grained interface layer; and the chip execution layer (4-4; 4-5), configured to call chip hardware modules to perform hardware acceleration according to the technical task and return the execution results to the upper layer. This invention thus achieves the technical advantage of “one-time development, multi-chip adaptation”.

The application provides a memory reuse method, a memory reuse device, equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: before the start of a first stage in an artificial intelligence model training process, determining whether first data corresponding to a second stage in the artificial intelligence model training process is transferred from a first memory of a computer to a second memory of the computer, the access speed of the first memory being higher than the access speed of the second memory, and the first data being data that will not be used in the first stage. The method further comprises the following steps: if it is determined that the first data has been transferred to the second memory, storing second data required by the first stage in a storage area in the first memory for storing the first data. The embodiments of the application can train an artificial intelligence model with a larger parameter scale in a predetermined storage space when training the artificial intelligence model, improve the utilization rate of high-bandwidth memory, and reduce the influence on the training performance.

Disclosed is a system (200) and a method (500) for managing hierarchical relationship among network entities in a computing environment. The system (200) detects removed (or status-changed) parent entity. Thereafter, the system (200) retrieves a policy configuration from a cache memory (215) and identifies child entities linked to the parent entity using the policy configuration. This enables faster and modifiable execution of operations without any overheads from the computing environment. The system (200), using the policy configuration, determines a stalling condition for the child entity and initiates stalling the child entity for a stalling period based on the stalling condition. The stalling of the child entity enables an option of reassociation with an active entity, based on a receipt of association request from the active entity, when attributes of the child entity match with a reassociation eligibility criteria defined by the policy governing the hierarchical relationship.

The present disclosure relates to a method and a system for communication between a microservice and a policy execution engine (PEEGN) The present disclosure encompasses receiving a request message comprising at least an event type and a publisher type, wherein the publisher type is at least one of the PEEGN and the microservice; determining a target publisher, associated with the request message, based on the publisher type, wherein the publisher type is the PEEGN; transmitting to the target publisher, the request message; receiving from the target publisher, a response for the request message, the response comprising a subscriber type associated with the request message; identifying a subscriber, associated with the request message, based on the subscriber type; an transmitting to the subscriber, the response associated with the request message.

A platform for facilitating an automated IT audit. The platform may have a frontend allowing users to access the platform, a backend configured to perform processing, and a data collection system equipped to interface with connectors. The backend may include at least one server equipped to send, receive, store, and process data; a testing and analyzing system that may make use of algorithms, machine learning, and artificial intelligence in order to test and analyze the collected data against pre-configured best practice standards and policies, and a reporting system that may be configured to transmit the tested and analyzed data to the frontend. The backend system may be configured to opine on the data and generate specific recommendations about future developments of an auditee's IT infrastructure, allowing an audit to be completed automatically from start to finish by the use of the software, eliminating the need for human intervention.

The application provides an event reporting method and device, a computer readable storage medium and a computer program product; the method comprises the following steps: receiving an abnormal event message sent by a target kernel module through an event bus, wherein the target kernel module sends the abnormal event message to the event bus when monitoring an abnormality, and the target kernel module is one of a plurality of kernel modules included in an operating system; sending the abnormal event message to a target monitoring service that subscribes to the abnormal event message, so that the target monitoring service analyzes and processes the abnormal event message after receiving the abnormal event message. Through the above method, the target kernel module and the monitoring service do not need to interact directly, the interaction cost of the kernel module and the monitoring service can be reduced, and the event reporting efficiency is improved.

The application discloses a multi-agent reinforcement learning method based on option-value decomposition, through a multi-agent option-value framework (OMIX), defines the action strategy of each agent as an option, the agent can select an option to imitate the action strategy of the corresponding agent, stops imitation through an option termination function, which can integrate other agent strategies, increase the exploration range and sample utilization rate; meanwhile, CTDE is used to realize multi-agent cooperation, global information is used in the training process, and is dispersedly arranged in the execution process; an optimistic heuristic method is integrated into option learning to prevent the agent from falling into local optimization in a sparse reward environment.

The specification provides a block generation method and device. The method is applied to a first node in a blockchain network, and the first node includes a cache process and a pre-execution process. The pre-execution process pre-executes a first transaction depending on block information of a block being packaged, and sends the pre-executed first transaction to the cache process. The cache process determines the block information depending on the pre-execution process of the first transaction, and determines first block information of a current block being packaged at the current time. If the block information is second block information different from the first block information, the cache process triggers the pre-execution process to re-pre-execute the first transaction depending on the first block information. If the block information is the first block information, the cache process packages the first transaction into the current block.

The application provides an extension method and device of a WEB service public component, electronic equipment and a storage medium. The public component includes a controller layer for parameter exchange with the outside world, a service layer for service under the call of the controller layer, and a data access layer for operating a database under the call of the service layer. The controller layer is not inheritable. The business logic of the public component is concentrated in the service layer. The method includes: creating an extension project dependent on the public component by depending on the service layer or the data access layer; deploying an extension service into the extension project by depending on the service layer; and configuring the extension project based on a startup class of the public component, so as to trigger the public component to provide the extension service based on the startup class of the public component. The application embodiment realizes zero-invasion strong extension of the WEB service public component.

The application relates to the technical field of computer software, and provides a message routing method, a system and electronic equipment, wherein the method comprises the following steps: firstly, after a message queue is accessed, current messages of the message queue are acquired, and description information of the current messages is determined; then, the description information of the current messages is compared with pre-configured rule information, a target processor for processing the current messages is determined, and the current messages are sent to the target processor. The routing method introduces the pre-configured rule information, and different access codes and consumption codes do not need to be written for different MQs, so that the consumption mode of the messages is no longer realized by hard coding, but realized by the configured rule information, so that the implementation cost of different MQs can be greatly reduced, the effective management capability for messages of different sources is improved, the application of the messages of different sources is facilitated, and the messages consumption hot replacement and hot suspension capability of a service with high restart cost can be realized.

The present invention provides a device, a computer-readable storage medium, and a method for exchanging data between containers, all equipped with shared memory. [Solution] Embodiments of this specification describe a Platform Abstraction Layer (PAL) for edge devices that enables services developed in different programming languages ​​to run as a pipeline on edge devices (e.g., cameras). For example, services may be deployed in separate containers developed using different programming languages. In one embodiment, stages in the pipeline can use buffers in shared memory to share data. The data may include a unique ID (e.g., frame ID) to identify the buffer containing the processed data by searching an index array for the shared memory. In this way, the shared memory supports random access.

The application provides a computational fluid dynamics simulation method and system and a computer readable storage medium, the method comprising: distributing storage space according to variable types, and loading to a shared memory of a calculation unit based on initial conditions; calculating a characteristic scale of the calculation unit, reducing a global time step, reconstructing a gradient combined with information of adjacent units, and calculating a flux; performing a preconditioning sub and a linear system operation in a matrix-free form to drive Krylov subspace iteration to update a solution vector, and repeatedly restarting iteration until convergence. The application solves the problem that the calculation efficiency of a large-scale system presents a nonlinear downward trend with the expansion of the scale, and it is difficult to fully utilize modern high-performance hardware computing power, and solves the defect that communication overhead generated by cross-node data transmission further restricts performance improvement of a synchronous operation, and optimizes resource utilization of a heterogeneous parallel architecture.

The application provides a time interval processing method and device, electronic equipment and storage medium, and relates to the technical field of data processing. The method comprises the following steps: in the case that a to-be-processed time interval is detected, a target set is acquired; the target set comprises at least one target time interval; if the to-be-processed time interval intersects with any target time interval, an original set corresponding to the target set is acquired; the target set is obtained by interval merging processing on the original set; the original set comprises at least one original time interval; the original set is updated based on the to-be-processed time interval to obtain an updated original set; and interval merging processing is performed on each original time interval in the updated original set to obtain an updated target set. The application can make the data quantity of the target set for managing and controlling time intervals lower than that of the original set, so that analysis or query of time management and control is performed based on the target set, and the efficiency is improved.

The application provides an audio time delay optimization method, a client, a device, a medium and a program product, and the method comprises the following steps: controlling a Native Module to acquire audio data; and controlling the Native Module to transmit the audio data to an audio hardware device in an MMAP memory mapping mode. The audio time delay optimization method provided by the application can directly transmit the audio data to a Native layer, avoids transmitting the audio data to a Java layer, and directly uses an underlying MMAP memory mapping mode in the Native layer to interact the audio data with the audio hardware device, so that the audio data does not need to be cached and copied layer by layer before reaching the audio hardware device, a shorter transmission path and less memory resource consumption can be obtained, the transmission efficiency of the audio data is improved, the sound quality of real-time audio and video office in a cloud desktop scenario is ensured, and the terminal load is reduced by saving more memory resource consumption.

Embodiments of the present disclosure relate to vehicle signals communicated using POSIX-based directory interfaces. A method of obtaining vehicle signal data. The method includes reading, by a communication service layer application executing on a computer system, a vehicle signal directory, wherein the directory defines vehicle signals as full path file names; receiving, by the communication service layer application, a request from a second application for a value of a vehicle signal, wherein the request identifies the vehicle signal as a full path file name; obtaining, by the communication service layer application, the value of the vehicle signal; and returning the value of the vehicle signal to the second application.

The application discloses a kind of automatic processing method, device and equipment of service request and storage medium, comprising: receiving the service data that data association service is fed back based on data extraction request;General service layer is specified to determine the non-data association service of pre-binding, and service data is sent to non-data association service;Receive the processing result of non-data association service for service data, and processing result is used as the service response of service request.Through the binding relationship of pre-binding service on general service layer and the application and connector on different function layer, when receiving user request, automation processing pipeline can be based on binding relationship, improve the processing efficiency and reliability of service request, and industrial application platform is divided into clear level, and the independence design of layer inner module makes each functional module highly cohesive and loosely coupled, so as to improve the processing efficiency and quality of service request.

The application relates to a cache task scheduling method and device, equipment, a storage medium and a program product. The method comprises the following steps: firstly, in the case of meeting a preset condition, obtaining the task priorities of cache tasks in a task pool comprising a plurality of cache tasks to be processed on a platform side; then, obtaining the network states of a plurality of cluster sides for processing the cache tasks on the platform side; then, determining a target cache task according to the network states of the cluster sides and the task priorities; and finally, performing task publishing processing on the target cache task, so that the target cache task is acquired and processed by the cluster sides. According to the method, the target cache task to be processed preferentially can be determined according to the task priorities of the cache tasks and the network states of the clusters, the cache task with high priority and good cluster network state can be processed more timely, and the scheduling of the cache task is more flexible.

Techniques are disclosed relating to a shared control bus for communicating between primary control circuitry and multiple distributed graphics processor units. In some embodiments, a set of multiple graphics processor units including at least first and second graphics processors on different semiconductor substrates that are packaged in a multi-chip module, where the first and second graphics processors are coupled to access graphics data via respective memory interfaces. The shared workload distribution bus may include: one or more interfaces between respective graphics processors on the same semiconductor substrate and at least one cross-substrate interface between the different semiconductor substrates. Workload distribution circuitry may transmit, via the shared workload distribution bus, control data that specifies graphics work distribution to the multiple graphics processor units. Packet control circuitry may modify packets from at least one of the one or more interfaces for transmission via the cross-substrate interface.

Embodiments of the present application provide a data transmission method, device, equipment, storage medium and program product. The method comprises: in response to a data sending request, converting, by a data converter, a data format of to-be-processed data into a target format corresponding to a target external system; if it is determined that authentication and authorization to the target external system are passed, sending, by a data transmission channel, the to-be-processed data in the target format to the target external system according to a data transmission mode of the target external system and an address of the target external system; in response to a data receiving request, if it is determined that the external system is in a white list, receiving the to-be-processed data; converting, by the data converter, a data format of the to-be-processed data into a target format corresponding to a target internal subsystem; and sending, according to interface basic information, the to-be-processed data in the target format to the target internal subsystem. The method is used to improve the efficiency of data transmission between a business system and an external system.

This invention discloses a method and processing system for improving the IO and concurrency performance of a ride-hailing pricing system. Applied to the server side, it receives and verifies intelligent pricing request data containing trip details, user information, and environmental supply and demand through an access interface; it performs pricing calculations based on business attributes to generate pricing result data; it writes this data to a local memory cache queue and synchronously records a write-ahead log (WAL) containing the cache queue status; immediately after writing to the cache, it returns the pricing result data to the request initiator, achieving asynchronous separation of response and persistence. The system monitors the cache queue status in real time and triggers a batch database entry process when a preset threshold is reached; it extracts multiple pricing data entries, groups them by ride demand identifier, cleans and deduplicates them, retaining only the pricing record with the latest timestamp to generate a set of pricing data to be entered into the database; finally, it writes the set to the database in batches and provides feedback on the persistence status through an asynchronous communication mechanism.

Systems, apparatus, articles of manufacture, and methods are disclosed to efficiently utilize computing resources for transferring data within computing systems. An example non-transitory machine readable storage medium comprising instructions to cause programmable circuitry to at least: determine a number of threads to prefetch data from a first matrix for matrix multiplication; assign a first plurality of threads to a first column of the first matrix to prefetch the first column into the first cache; assign a second plurality of threads to a second column of the first matrix to prefetch the second column into the second cache; and assign a subset of the first plurality of threads and the second plurality of threads to calculate products of the first matrix from the first and second cache and a second matrix.