2230results about "Program synchronisation" patented technology

The invention relates to the field of data backup, in particular to an intelligent data backup method and system based on an AI large model. The method comprises the following steps: acquiring an enterprise global data list, performing intelligent data structure deconstruction and dynamic attribute mapping modeling, and constructing a holographic data semantic perception model; performing real-time transient risk mutation detection on the holographic data semantic perception model, and constructing an intelligent backup triggering mechanism; carrying out storage resource demand prediction based on an intelligent backup trigger mechanism, carrying out multi-storage cloud environment resource dynamic scheduling, and constructing an elastic backup storage resource pool; carrying out incremental backup analysis and self-adaptive compression coding to obtain an incremental backup coding packet; and performing dynamic backup sequence adjustment and intelligent incremental backup decision on the incremental backup coding packet based on the elastic backup storage resource pool, and constructing an intelligent incremental backup execution engine. According to the method, the reliability, the accuracy and the traceability of a backup result are improved through self-adaptive intelligent incremental backup.

A parallel processing unit comprises a plurality of processors each being coupled to a memory access hardware circuitry. Each memory access hardware circuitry is configured to receive, from the coupled processor, a memory access request specifying a coordinate of a multidimensional data structure, wherein the memory access hardware circuit is one of a plurality of memory access circuitry each coupled to a respective one of the processors; and, in response to the memory access request, translate the coordinate of the multidimensional data structure into plural memory addresses for the multidimensional data structure and using the plural memory addresses, asynchronously transfer at least a portion of the multidimensional data structure for processing by at least the coupled processor. The memory locations may be in the shared memory of the coupled processor and / or an external memory.

The invention discloses an android container rendering optimization method based on cross-domain hard real-time Fen synchronization, which comprises the following steps: by taking a swan-mong system as a host and an android system as a container, creating a hash table, a synchronous thread and a shared memory corresponding to a GPU core when the host is started, acquiring VSync cycle registration callback, transmitting shared memory FD to the container, and finishing shared memory mapping and alignment by the container. Registering a GPU queue to complete callback; when the Android application is started, a container obtains a queue and a physical address of a rendering buffer area, creates a Fen and binds the Fen to the queue, after the queue is submitted, metadata is written into a shared memory to inform a host, after the host receives the metadata, nodes are created and stored in a hash table, and an overtime timer is registered; after the GPU completes the command queue, the container calls back an update state and a verification value to notify the host, and after the host is verified to be valid, the corresponding hash table is updated, the timer is reset, and asynchronous screen loading is triggered; and the host executes buffer area synthesis and submission of the display equipment to finish on-screen, so that the stability of the rendering frame rate is improved, and the reliability of cross-domain synchronization is ensured.

Apparatuses, systems, and techniques to perform a neural network. In at least one embodiment, an application programming interface schedules two or more graph nodes to be performed by two or more parallel processing pipelines based, at least in part, on an order of layers in the neural network.

The invention provides a multi-transaction deadlock detection and processing method in resource-constrained equipment, which comprises the following steps of: in a transaction execution process, constructing a resource waiting graph which is used for recording a waiting relationship among transactions so as to reflect the occupation condition of shared resources by the transactions and a dependency relationship among the transactions; before a transaction is subjected to read-write operation each time, whether execution of the current transaction can cause a deadlock risk or not is judged in real time by detecting a dependency relationship between the current transaction and a target resource and combining a resource waiting graph; and when a deadlock risk is detected, deadlock avoidance strategies of priority processing, transaction rollback, transaction retry and exception processing are adopted to avoid deadlock. The deadlock problem occurring in the transaction parallel processing process can be effectively avoided or solved, and meanwhile the response efficiency of the application and the utilization rate of the storage space are guaranteed.

A new transaction barrier synchronization primitive enables executing threads and asynchronous transactions to synchronize across parallel processors. The asynchronous transactions may include transactions resulting from, for example, hardware data movement units such as direct memory units, etc. A hardware synchronization circuit may provide for the synchronization primitive to be stored in a cache memory so that barrier operations may be accelerated by the circuit. A new wait mechanism reduces software overhead associated with waiting on a barrier.

The invention provides an online document processing method and system based on a document processing model, and the method comprises the steps: obtaining an editing request of a user for a target data unit, generating a temporary locking identifier according to a user authority identifier and a request timestamp, limiting the modification operation of other users for the target data unit, and obtaining a conflict isolation result; constructing a data consistency module according to a conflict isolation result, extracting editing state information of the current data unit from the temporary locking identifier, and performing sequential recording on concurrent operation by adopting a distributed transaction processing technology; if it is detected that version conflicts exist in the sequenced records, a rollback mechanism is triggered, operation priorities are redistributed, and data versions after consistency verification are determined; and designing a hierarchical cache structure according to the data version after consistency verification, storing the high-frequency access data into a memory cache region, and storing the low-frequency access data into a disk cache region to obtain a hierarchical storage structure. The data consistency and security are good, and the storage and access efficiency is improved.

The invention discloses an information resource efficient collaborative optimization method for a distributed computer system, which belongs to the technical field of information resource optimization control and comprises the following steps: executing a transaction according to a target lock element information set, monitoring popularity information of each resource segment, generating self-adaptive lock granularity decision information of each resource segment, and executing dynamic adjustment of lock granularity. And obtaining load sensing optimization parameters to complete lock competition optimization, and completing real-time iterative updating of lock element information and strategies. According to the method, the problems of resource competition and deadlock in the distributed system can be effectively solved, and the system performance and the resource utilization efficiency are improved. Transaction pre-allocation resources are reasonably allocated, the deadlock risk is reduced, lock granularity is dynamically adjusted according to resource segmentation popularity, lock competition optimization is completed based on load awareness optimization parameters, an adjustment result and the optimization parameters are synchronized to a distributed lock management library for real-time iteration updating, and it is ensured that a system always operates with an optimal strategy; and the distributed computer system efficiently and stably processes information resources in a cooperative manner.

The invention relates to the technical field of data processing systems, and discloses a command and dispatch management method and system based on artificial intelligence, and the method comprises the steps: generating a space-time occupancy statement containing an occupancy area, a time window and a task state through an edge node, and broadcasting the space-time occupancy statement to an adjacent node; and a receiver performs local conflict verification according to a state mutual exclusion rule, and triggers priority-based lightweight negotiation to adjust an occupied time window when detecting that static observation and dynamic intervention states coexist in an overlapped space-time. Postward conflict processing of traditional centralized scheduling is converted into beforehand conflict avoidance, the response speed and the operation toughness of the system are remarkably improved, and meanwhile, through physical reachability verification and a channel self-adaption mechanism, the cooperation reliability in a complex environment is guaranteed.

The invention discloses a cloud server physical GPU acceleration capability release method and system, and belongs to the technical field of computer GPU capability release, and the method specifically comprises the steps: collecting GPU operation parameters in real time, carrying out the preprocessing, storing the GPU operation parameters in a shared state database, constructing a lightweight virtualization abstraction layer on a physical GPU, dividing the physical GPU into a plurality of logic units, and storing the logic units in a shared state database; the method comprises the following steps: establishing a physical GPU, constructing an isolation layer, utilizing a three-dimensional resource segmentation strategy, adopting a priority queue, a jump mechanism and a delay tolerance mechanism in a scheduling process, dynamically allocating logic unit resources of the physical GPU, accelerating release of GPU resources, and realizing unification and real-time updating of physical GPU state information among nodes of a cloud server through a distributed state synchronization mechanism; through dynamic task scheduling, a jumping mechanism, a delay tolerance mechanism and distributed state synchronization, efficient GPU resource management is realized, and GPU capability release is accelerated.

The invention discloses a workflow-based multi-modal water conservancy large model decision support method, which comprises the following steps of: S1, preprocessing multi-modal input data based on a local model, and converting a data format; s2, constructing a modular workflow supporting parallel processing through a specified DSL (Digital Subscriber Line) framework; s3, constructing a feature model suitable for the water conservancy field through the multi-modal data subjected to synchronous acquisition and processing, and performing field adaptation and fine adjustment on the feature model; s4, on the basis of the feature model subjected to domain adaptation and fine adjustment, a specified workflow is constructed, and multi-modal task resources are dynamically scheduled; and S5, combining a local knowledge base with networking data by using a dynamic fusion algorithm to realize dual-channel knowledge fusion for executing the multi-modal task in the S4 in parallel. According to the invention, real-time acquisition and parallel processing of multi-modal data can be realized, and the stability of a local knowledge base and the real-time performance of networking data are dynamically balanced through a dual-channel knowledge fusion mechanism.

The invention provides a write-after-read conflict prediction method, device and equipment, which are applied to the technical field of data processing, and the method comprises the following steps: reading a to-be-transmitted loading instruction; the loading instruction is located in a loading queue of the out-of-order execution processor. Determining an index value of the loading instruction; the index value is used for quickly positioning a conflict record matched with the loading instruction in a historical conflict table of the out-of-order execution processor. The conflict record comprises a first storage instruction corresponding to the loading instruction. And when it is determined that the conflict record corresponding to the index value exists in the historical conflict table, setting the loading instruction to be in a suspended state. And after the execution of the first storage instruction is finished, executing the loading instruction. The problem that the execution efficiency of a processor is affected due to pipeline emptying and pause caused by write conflicts after reading can be solved.

The invention provides a distributed event processing method and electronic equipment, which can be applied to the field of distributed technologies. The distributed event processing method comprises the following steps: receiving an event processing request at a request receiving entrance of a distributed system, and searching a priority and a mutual exclusion execution type corresponding to an event type from a preset event registration library according to an identifier representing the event type in the event processing request; based on the priority and the timestamp of the event processing request, inserting an event object obtained after packaging the event processing request into a priority scheduling queue; after an event object used for execution is extracted from the priority scheduling queue, conflict detection is carried out on a mutual exclusion execution type corresponding to the event type based on a global mapping table maintained in the distributed system; and under the condition that the conflict is not detected, calling a service processor corresponding to the event type to execute the event object, and updating the current execution count of the event type in the global mapping table based on the execution state.

The invention relates to the technical field of agent assessment, and discloses an AGENT equipment health assessment system and method based on a knowledge graph, and the method comprises the following steps: constructing a resource topological graph, loading equipment state and resource capacity data, and providing a structured knowledge basis for collaborative assessment; the equipment state comprises a health degree and a priority; quantizing the equipment resource demand, generating an equipment-resource demand matrix, and determining the dependency relationship of each piece of equipment on each type of resources; and the maintenance agent predicts the equipment resource demand, the resource agent performs deadlock detection in combination with the demand and the resource constraint, and a deadlock risk probability and a fault equipment list are output. According to the invention, through collaborative optimization of the maintenance agent and the resource agent, generation of the dynamic scheduling strategy, real-time risk monitoring and rollback mechanism triggering, and formation of a closed-loop knowledge graph updating mechanism, the system has a sustainable evolution adaptive capability, the operation and maintenance cost is effectively reduced, and faults caused by system decision conflicts are prevented.

A method for managing a data stream processing pipeline includes: monitoring, by an orchestrator, data stream ingestion of a streaming storage system (SSS) to obtain data stream metrics; analyzing, by the orchestrator, the data stream metrics based on a user-defined scaling policy; making, based on the analyzing and by the orchestrator, a first determination that task manager scaling is required; making, by the orchestrator, a second determination that the data stream ingestion is increased, in which the second determination indicates an increase in a number of parallel stream segments associated with a data stream, in which a segment store hosted by the SSS manages the parallel stream segments; and to increase a stream processing system's compute capability and by the orchestrator, initiating scaling of a number of the plurality of task managers to support the increase in the number of the parallel stream segments.

Embodiments of the present application provide a transaction processing method, a device, and a storage medium. For a distributed transaction, it is proposed innovatively that exclusive locks are added to data records involved in a read set of the distributed transaction in a read phase; in a write phase, write operations for a write set can be directly executed based on these exclusive locks, and an exclusive lock on a related data record can be released immediately after a write operation is completed. In the write phase, the exclusive locks added to the data records can maintain the exclusivity of the distributed transaction over these data records, which can avoid the write operation abort problem in the write phase. That is, all write operations of the distributed transaction are bound to succeed, thereby ensuring the atomicity of distributed transaction processing. In addition, since the exclusive lock is released immediately after the write operation is completed, the lock-holding time of the distributed transaction is only the time inherently required to execute transaction logic of the distributed transaction, thereby greatly reducing the lock-holding time of the distributed transaction, effectively increasing the throughput for distributed transactions and improving the transaction processing efficiency.

The invention relates to a synchronous processing method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of obtaining a to-be-processed task, splitting the to-be-processed task into at least one to-be-processed instruction, and distributing the to-be-processed instructions to different thread processing units; obtaining an access sequence barrier corresponding to the to-be-processed task, and sending the access sequence barrier to each thread processing unit; when the execution of the at least one to-be-processed instruction is completed, generating access sequence barrier execution completion feedback information, and storing the access sequence barrier execution completion feedback information to a target barrier cache region corresponding to the thread processing unit; if the corresponding access sequence barrier execution completion feedback information is stored in each target barrier cache region, it is determined that the to-be-processed task is completed, and the target barrier cache regions correspond to the thread processing units. By adopting the method, the processing efficiency can be improved.

The invention provides a dynamic tool integration method and system for an intention-driven AI dialogue system, and the method comprises the steps: S5, extracting resource occupation information from a transaction log according to a state persistence record, carrying out the resource release and temporary context cleaning of each heterogeneous component, and obtaining a released resource list; s6, searching a breakpoint state snapshot and a transaction log from the distributed database according to the released resource list, and reconstructing a tool chain execution flow historical state to obtain a fault context analysis result; and S8, according to the context recovery state and the semantic feature vector, updating a tool chain execution stream, and reallocating computing resources through a heterogeneous component coordination mechanism to obtain a task continuity confirmation result. According to the method, the recovery efficiency and the resource utilization rate of the tool chain in an abnormal scene are remarkably improved, and the task continuity and the system robustness in a high-concurrency environment are guaranteed.

The invention relates to the technical field of computers, in particular to a dynamic configuration initialization and version synchronization method and system for multi-agent service, and the method comprises the following steps: obtaining a database update lock, constructing a database migration script dependency relationship, determining a current version of a database mode, upgrading and deploying the database mode, and verifying agent versions. The method has the beneficial effects that a rigorous database updating lock control mechanism, perfect migration script dependency chain verification, agent version management, port availability detection, idempotence and transaction safeguard measures are introduced, so that the processes of database structure upgrading and agent version initialization in a distributed deployment environment can be effectively supported; and the data consistency, the operation stability and the service continuity of the system are ensured.

The invention relates to the technical field of storage management, and discloses a processing resource allocation method and computing device.The method comprises the steps that a memory allocation request of a first process to a second processor is intercepted, and the memory allocation request is used for requesting the second processor to allocate a storage area of a first storage size for the first process; determining a lock holding state of the first process to a dynamic time slice DTL lock; and under the condition that the first process holds the DTL lock, calling the second processor based on the memory allocation request, so that the second processor allocates a first storage area in the memory of the second processor for the first process. Therefore, process scheduling can be carried out through the DTL lock, and the storage area is managed under the condition that the process holds the DTL lock, so that frequent page immigration and emigration are avoided, and the jitter phenomenon is reduced.

The invention relates to the technical field of graphics processors, and provides a dynamic graphics processor sharing method, which comprises the following steps of: receiving submission of a plurality of deep learning tasks; generating a dynamic priority queue containing high-priority tasks and low-priority tasks; transmitting a resource state to the low-priority task through a broadcast mechanism, and triggering resource adjustment according to a comparison result of the real-time thread block usage amount of the high-priority task and a preset threshold value; according to the difference value between the real-time thread block usage amount of the high-priority task and a preset threshold value, the number of executable thread blocks of the low-priority task is dynamically adjusted; checking whether the number of thread blocks required by the kernel is smaller than or equal to the number of executable thread blocks of the current low-priority task, if yes, starting execution, and if not, reserving the kernel in a kernel request queue to wait for next cycle scheduling; and periodically updating the dynamic priority queue until all deep learning tasks are completed. According to the technical scheme, the overall efficiency of the system can be improved, and resource waste is reduced.

A computer-implemented method of distributed management of locks includes: receiving, by a plurality of servers, a plurality of concurrent requests; performing lock preemption on the distributed lock using a synchronization mechanism; only one server uses a synchronization mechanism to obtain the distributed lock; updating the lock value of the distributed lock by only one server; and executing the business logic. Lock preemption is performed on the distributed lock by the respective server, including comparing a current memory value of the distributed lock to an expected value. The method further comprises the steps that if the current memory value is matched with the expected value, only one server modifies the current memory value into an updated value; writing the updated value into a memory of the distributed lock; and if the execution of the service logic fails, recovering the memory value of the distributed lock to the current memory value.

The invention discloses an order processing method and system based on a message queue, and belongs to the technical field of computers, and the method comprises the steps: receiving order data from a business platform, carrying out the preprocessing of the order data, reading related fields from a preset business rule library, and carrying out the data marking of an order; intelligently distributing orders by using a rule engine, and pushing integral messages of the distributed orders to corresponding consumption queues through message queues; after the order adding and inventory reserving operation is successfully submitted, submitting a subsequent state flow operation of the order as an asynchronous task to a state flow queue for processing; key indexes are collected through the real-time monitoring module, monitoring data are fed back to the dynamic rule module, and self-adaptive optimization of an order distribution strategy is achieved. According to the method, the row lock contention of the database can be remarkably reduced, the concurrent processing capability of the system is improved, the system stability and the data consistency are enhanced, and the resource utilization and the system expansibility are optimized.

The invention provides a task synchronization method and device, electronic equipment and a storage medium. The method comprises the following steps: a source chip and a target chip respectively receive task packets carrying unique identifiers; the source chip randomly allocates a first barrier identifier for the task packet, and establishes a local mapping table of the unique identifier and the first barrier identifier; the target chip randomly allocates a second barrier identifier for the same task packet and establishes a local mapping table of the unique identifier and the second barrier identifier, and the first barrier identifier and the second barrier identifier are mutually independent; after finishing local operation, the target chip sends a synchronization ready signal containing a unique identifier to the source chip through the inter-chip communication path; the source chip queries a local mapping table according to the unique identifier, obtains a corresponding first barrier identifier and triggers a synchronization signal; and when the source chip counts that the number of the synchronization signals of the first barrier identifier meets the preset condition, the cross-chip task is executed, so that the task synchronization efficiency and the parallel computing capability of the cross-chip task are improved.

One embodiment provides a multi-chiplet graphics processor comprising a plurality of chiplets, where a chiplet of the plurality of chiplets comprise a memory interface, processing resources configured to execute threads of a kernel, and thread dispatch circuitry to facilitate dispatch of threads of the kernel to the processing resources. The processing resources are configured to execute threads of a first kernel, receive dispatch of threads of a second kernel for execution before completion of the first kernel as threads of the first kernel retire, execute a first phase of the second kernel during completion of execution of the first kernel, via a thread of the first kernel, signal an event via an uncached write to a global memory, and execute a second phase of the second kernel based on detection of the event via an uncached read from the global memory.

Techniques for bus sharing are disclosed. A plurality of bus requesters is coupled to a common bus. Access requests are stored in a request vector. A relative request vector (RRV) is generated by circular shifting by a rotate value to the right. The RRV is examined for the first asserted bit beginning at a first bit position proceeding in a first direction. A one-hot encoded relative grant vector is generated based on the first asserted bit. A grant vector (GV) is produced by circular shifting by a rotate value to the left. Access is granted. The rotate value is updated, affecting the priority of next grant. The GV is compared to a grant history vector (GHV). A count grant is adjusted. The count grant indicates a number of times that the bus requester indicated by the GHV was granted successive accesses to the common bus.

The invention relates to the technical field of multi-agent collaboration, in particular to a multi-agent collaboration method and system based on shared memory data exchange, and the method comprises the steps: creating and initializing a shared data area which is mapped and accessed by a plurality of agent processes in a physical memory; constructing an annular buffer structure on the shared data area, and realizing a lock-free read-write pointer propulsion mechanism based on atomic operation and a memory barrier so as to support a plurality of intelligent agents to perform data read-write concurrently; asynchronous notification of data updating is carried out between agent processes through a lightweight event notification mechanism; and each agent performs autonomous decision making and task scheduling based on the shared data so as to realize distributed collaboration. A data block version control and verification mechanism is introduced into the annular buffer structure, and when it is detected that the agents are abnormal, resources of the abnormal agents are automatically recycled, and a data consistency recovery process is triggered. Multiple times of data copying between a kernel mode and a user mode are eliminated, and communication delay is reduced.

Skills and skills metadata may be used to define a process for building a data center. Skills of one service may depend on skills corresponding to the same or different service. A dependency graph may be generated based on these dependencies. The graph may specify an order by which orchestration operations are to be performed to build the services, thereby building the data center. During execution of the process for building the data center, health states corresponding to the skills may be tracked (based at least in part on alarms and / or namespaces associated with the skills). When an unhealthy skill is identified, the system may traverse the dependency graph to identify a root cause (e.g., failed operations corresponding to a skill on which the unhealthy skill directly / indirectly depends). A notification and / or various options may be provided to address the unhealthy state of one or both skills.

The invention provides a sparse approximate inverse Shenwei parallel preprocessing method and system for electromagnetic scattering, and relates to the technical field of processor parallel computing, and the method comprises the steps: carrying out the performance hotspot analysis of a sparse approximate inverse preprocessor in an electromagnetic scattering simulation process, and determining a hotspot function of the sparse approximate inverse preprocessor; extracting a sparse matrix vector multiplication operation of a hotspot function, performing master-slave core parallel calculation on the sparse matrix vector multiplication operation, migrating the sparse matrix vector multiplication operation to slave core calculation, and accelerating hotspot calculation by using a heterogeneous parallel sparse approximate inverse algorithm; parallel computing of master-slave core intensive numerical computing tasks is achieved; in the parallel computing process, the slave cores communicate with the master core in a direct memory access mode, and during the period, the master core is in a waiting state to ensure data consistency and communication synchronization until all the slave cores compute distributed tasks.

The invention provides a parallel world simulation resource dynamic scheduling optimization method and system. The method comprises the following steps: detecting a dependency relationship between tasks in parallel world simulation resources to obtain topological data, and quantifying resource competition intensity and sequence constraint to form an associated feature matrix; a cross-dimension optical signal transmission path is constructed through dynamic change of a programmable photon array analysis matrix, and a low-conflict resource transmission channel map is formed; synchronously processing the matrix and the channel map to generate a global task execution sequence; executing a multi-world resource occupancy operation and generating a resource occupancy relation graph; and calculating a cross-dimension resource conflict probability in the atlas, and dynamically adjusting a task execution sequence to avoid a simulation deadlock risk. According to the method, nanosecond-level dynamic resolution of cross-dimension resource conflicts is realized, the simulation deadlock rate is zeroed, and the task throughput is improved.