1456 results about "Shared memory" patented technology

A method to share remote DMA (RDMA) pointers to a key-value store among a plurality of clients. The method allocates a shared memory and accesses the key-value store with a key from a client and receives an information from the key-value store. The method further generates a RDMA pointer from the information, maps the key to a location in the shared memory, and generates a RDMA pointer record at the location. The method further stores the RDMA pointer and the key in the RDMA pointer record and shares the RDMA pointer record among the plurality of clients.

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 a cross-container application fusion switching method of a swan monk system, which comprises the following steps: a system service layer deploys a container application management service, an application framework layer realizes a proxy application manager, when a container application is started, a container side allocates a shared memory and configures authority, the container manager collects metadata to initiate registration, and the application framework layer realizes a proxy application manager; the container application management service converts a memory handle into a texture handle, allocates a unique identifier and triggers a proxy application manager to generate a proxy application, and the proxy application initializes a Vulkan rendering environment; the container side renders an application interface to a shared memory, the proxy application imports texture and constructs a lightweight Vulkan rendering pipeline, and scaling sampling is carried out to generate a thumbnail of the container application; and when the container application exits, the shared memory is released, the container application management service cleans the shared memory reference, triggers and destroys the proxy application, and recycles the texture resources through the reference counter, so that seamless fusion, low-delay switching and efficient resource utilization of the container application in a native application thumbnail form are realized.

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.

The invention discloses a container application on-screen method based on texture full-process optimization, which comprises the following steps: taking a swan-gap system as a host and an Android system as a container, distributing a plurality of first buffers by the host, screening out a first texture format by the container, writing original textures of the format generated by rendering into frame buffers, identifying a dirty area of the rendered textures, and displaying the original textures in the frame buffers; packaging the texture data of the region, storing the packaged texture data into a shared memory, transmitting a file descriptor FD to a host, binding an idle first buffer setting priority, a state, a generation timestamp and a texture data pointer at the same time, and obtaining the texture data by the host through the FD; the host establishes a fixed thread task pool, executes multiple tasks in parallel to obtain on-screen texture data, and then updates the state of the first buffer and a texture data pointer; and finally, the host queries the first buffer with the processed state, reads the texture data after sorting according to the priorities and the generation timestamps to finish on-screen, and resets the state of the first buffer to be idle, so that the on-screen delay is effectively reduced, and the actual frame rate of the Android application is improved.

The invention discloses a network card data local preprocessing system fused with edge computing, and relates to the technical field of edge computing and artificial intelligence collaborative optimization. Comprising an edge computing unit, a hierarchical collaborative architecture, a model hot switching and generative fragmentation module, an intention recognition and adaptive scheduling module, a delay energy consumption optimization scheduling module, a CXL zero-copy sharing module, an edge computing unit integrated processor, an FPGA or ASIC and a neuromorphic computing unit. According to the method, an FPGA, an ASIC and a neuromorphic computing unit are integrated in an intelligent network card, microsecond-level dynamic connection reconfiguration and adaptive generative model fragmentation execution are realized through a reconfigurable Mesh interconnection matrix, an attention layer and a feed-forward layer of a Transform class model are fragmented and allocated to different computing units for parallel execution, and cross-card streamlined processing is realized in cooperation with a zero-copy shared memory. And the intention recognition module is deeply coupled with the model hot switching module, so that dynamic model switching and fragmentation strategy optimization based on service priorities and system loads are realized.

The invention discloses an attention mechanism calculation method and device, a storage medium and a product, and the method comprises the steps: carrying out matrix multiplication operation through employing a query matrix block of a first register block and a key matrix block of a shared memory, obtaining a first product matrix block, and writing the first product matrix block into a second register block; performing exponential operation by using the first product matrix blocks to obtain sub-matrix blocks, writing the sub-matrix blocks into a second register block in a covering manner, and writing the sub-matrix blocks into a third register block in the form of a target precision type; performing matrix multiplication operation by using the sub-matrix blocks of the third register block and the value matrix blocks of the shared memory to obtain second product matrix blocks, and writing the second product matrix blocks into a second register block; performing softmax operation by using the second product matrix blocks to obtain attention result matrix blocks, and writing the attention result matrix blocks into a fourth register block; and writing the attention result matrix of the fourth register group into the shared memory in blocks. According to the embodiment of the invention, overflow of the register can be avoided, and the utilization of hardware resources is maximized.

The invention discloses an attention calculation implementation method and device, a medium, equipment and a product, and the method comprises the steps: loading the current to-be-calculated ith query block from a shared memory to a first register group of a consumer thread group, and carrying out the internal splitting calculation of the query block and a key block, so as to obtain corresponding attention score blocks; sequentially obtaining MK attention score blocks, and executing attention fusion calculation of mixing precision with the corresponding value blocks to obtain an attention output block corresponding to the ith query block; and after the attention output blocks are logically divided into N2 batches, a specified register group for storing target precision type data in the fusion calculation process is multiplexed and executed according to batches, target precision type conversion is carried out, and an output result after conversion of each batch is written back to a shared memory. According to the method, the existing hardware resources can be efficiently utilized to improve the calculation performance, and the method is particularly suitable for large-size query block and key block scenes.

A metadata access method, apparatus, and computer-readable storage medium for efficient metadata retrieval through cache management. The method receives metadata query requests including target index information from processes and performs matching operations on a global cache file containing records with index information and slot identifiers. Each cache description array corresponds to memory blocks caching metadata. Upon successful matching, the target cache description array is accessed using the target slot identifier. The data state of target metadata is determined from the cache description array, and target address information indicating the location of the target memory block in shared memory is obtained and returned to the requesting process, enabling efficient shared memory-based metadata access.

The invention discloses an MQTT message transmission optimization method and system. The method comprises the steps of analyzing a theme, extracting a device type, a data feature and a geographic position triple, calculating a hash value, and mapping a device to a specified Broker fragment cluster node to generate a fragment mapping table; processing the equipment data in the edge domain in the fragment mapping table through an edge calculation layer, removing invalid data according to a preset rule, merging the equipment data in the same fragment node, and embedding a fragment node ID for an aggregation message generated after merging; identifying a fragment node ID and routing to a target fragment node, positioning a corresponding shared memory pool, and writing the message into the shared memory pool; and responding to a direct access request of the client, so that the client directly accesses the data from the shared memory pool through the user mode network stack. According to the method, the problem of uneven load is effectively solved, multiple times of state switching in the data transmission process is avoided, and the MQTT message transmission efficiency is improved while the transmission cost is reduced.

The invention discloses an access method, device and equipment and a computer readable storage medium, which are applied to the technical field of computers, and comprise the following steps: loading a consistency label list corresponding to an authorized access consistency domain; when an access request is initiated to a target consistency domain in the shared memory, performing permission verification on the access request through the consistency label list and the target consistency domain; when the permission verification is passed, determining whether an access conflict exists or not through a local access state cache and the type of the access request; if the access conflict exists, a controller arbitration engine is triggered, and the access request is executed according to an arbitration result; and if the access conflict does not exist, executing the access request. According to the method, the consistency label list and the access state cache are formulated for each device, the consistency problem existing when multiple devices access the shared memory in an existing system is solved, and efficient, low-conflict and flexible consistency guarantee can be achieved under the scene that multiple devices concurrently access the shared memory.

The invention relates to a virtual network card implementation method based on a shared memory, an electronic device and a medium, which take the shared memory as a core communication carrier, construct a low-overhead data transmission basis, and efficiently implement a producer-consumer model in combination with the cyclic storage characteristic of an annular buffer area and the lock-free synchronization mechanism of atomic operation. Through partition design and atomic operation of the annular buffer area, continuity and exclusiveness of data access are guaranteed, and performance loss of a traditional lock mechanism is avoided; meanwhile, on the basis of the IPI interrupt capacity of the universal interrupt controller, cross-core synchronous notification is achieved through MPIDR accurate routing, interrupt wakeup waiting tasks are triggered through hardware, low-efficiency polling is replaced, and high-frequency data interaction scenes such as virtual network equipment are perfectly adapted. Compared with the prior art, the method has the advantages of efficient, safe and reliable network communication and the like.

The invention provides a BLAS3 structured operator accelerated computing system based on a Hopper architecture GPU, and relates to the technical field of computers. The system comprises: a calculation unit discrimination module for determining a calculation unit used by a current operator during operation, and estimating the maximum row dimension upper bound of the current operator in a tensor core execution path; an instruction sensing block parameter determination module dynamically determines the optimal block size and number of the input matrix in real time; the block matrix loading and aligning module divides an input matrix and a matrix to be updated into sub-matrixes by taking the block size as a basic block and completes loading of the corresponding sub-matrixes; the operator kernel function execution module completes shared memory structured parallel loading and storage of a double-precision floating-point number array of a sub-matrix corresponding to the input matrix, and calls a tensor core to carry out multiply-add accumulation calculation; and the assembly line and concurrent scheduling module adds the block calculation tasks into corresponding task sets and performs multi-stream concurrent scheduling on the task sets.

The invention discloses a Linux container-based Android and open source gap dual-system collaborative operation method, relates to the technical field of operation system collaboration, and aims to reduce resource occupation and improve collaboration and switching coherence, and the method comprises the following steps: step 1, by taking an Android as a host, creating a container isolation running domain, starting a namespace and managing resources by a control group; the method comprises the following steps: presetting a binder virtualization and three-state template; 2, deploying an open source gap in a container, mounting a shared directory in a directory binding manner, configuring bridging or address translation, and establishing copy-on-write and snapshot identifiers; and step 3, establishing a cross-system binder channel, and setting a service registration agent in a service manager and completing mapping of a handle and an authority domain by combining a shared memory and a local socket. And step 4, executing system switching without restarting, enabling one foreground and the other background through a foreground interface and continuous quota assignment, freezing a background session, limiting the speed of background heartbeat, and returning according to a snapshot when an exception occurs.

The invention relates to an APP intelligent marketing service method based on intelligent routing and multi-agent cooperation. The method comprises the following steps: receiving input information, wherein the input information comprises user active inquiry information or trigger event information generated based on user behavior monitoring; semantic analysis and intention recognition are carried out on the input information, a task planning directed acyclic graph is generated based on a recognition result, and the task planning directed acyclic graph comprises a plurality of subtask nodes and dependency relationships among the nodes; based on the task planning directed acyclic graph, the state of each functional agent and the historical performance index, executing dynamic routing so as to dispatch the plurality of sub-tasks to the corresponding functional agents; and in the execution process of the plurality of subtasks, performing dependency scheduling and state consistency management on an external tool call chain which is initiated by the functional agent and comprises a plurality of steps, and updating the shared memory associated with the user based on an execution result. By adopting the method, self-adaptive planning, robust execution and continuous optimization of marketing tasks can be realized.

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.

The invention relates to the technical field of computers, discloses a heterogeneous computing low-delay communication method and system, and aims to solve the problem of high delay caused by high communication protocol overhead, lack of dynamic scheduling collaboration, memory migration redundancy and non-uniform cross-node communication abstraction in existing heterogeneous computing. The method comprises the following steps: receiving a task scheduling request and analyzing a task dependency graph; tasks are dynamically allocated based on node loads and link states; rDMA, NVLink or PCIe straight-through protocols are adaptively selected according to node types to establish communication channels; zero-copy data exchange is realized through a shared memory mapping buffer area; hardware timestamps are utilized to synchronize feedback delays with PTP to optimize scheduling. The system comprises a heterogeneous computing node cluster, a unified communication scheduling controller, a low-delay communication protocol stack, a shared memory mapping buffer area and a communication delay sensing task distributor. According to the scheme, the communication delay is remarkably reduced, and the throughput and the task execution efficiency are improved.

The invention belongs to the field of artificial intelligence and knowledge graph application, and relates to an intent-driven large model knowledge graph intelligent question and answer method and an intelligent agent. The method comprises the steps of receiving a question input by a user, and identifying a potential intention of the user; according to the identified intention and the information stored in the shared memory, selecting a proper tool from a tool only based on a large language model, a tool based on semantic analysis, a tool based on sub-graph retrieval and a general tool for reasoning; new information generated by the tool is updated into the shared memory; and in each iteration, the large language model evaluates the current reasoning state according to the information stored in the shared memory, and triggers a tool with a corresponding intention to perform supplementary reasoning according to the current reasoning state until enough information is collected to draw a conclusion or a preset upper limit of the number of iterations is reached. The method can improve the reasoning efficiency and accuracy, can be applied to complex question answering in various fields, and provides high-quality question answering service for users.

The invention relates to a customer service scene-oriented collaboration system and method, and belongs to the technical field of artificial intelligence. In the system, a shared memory storage center stores knowledge entries from a plurality of customer service agents in a predefined data structure; the shared memory storage center is provided with an access control mechanism associated with a visible domain identifier, and an access range is limited during retrieval through a logic partition; the memory management module receives candidate knowledge entries from each customer service agent based on a trigger event, determines visible domain identifiers for the candidate knowledge entries, associates the candidate knowledge entries with the visible domain identifiers, and stores the associated candidate knowledge entries and the visible domain identifiers in the shared memory storage center; the problem processing and decision-making module responds to an access request from any customer service agent and determines a corresponding visible domain identifier according to the identifier of the customer service agent; and executing access based on the visible domain identification and the input question, and retrieving knowledge entries with access from a shared memory storage center to realize cross-agent multiplexing.

The invention discloses a multi-mode underwater acoustic communication system and communication method based on a heterogeneous dual-core architecture, and belongs to the technical field of underwater acoustic communication. The multi-mode underwater acoustic communication system comprises a main control module, an acquisition module and a transmitting module; the main control module adopts a dual-core heterogeneous processor OMAPL138 formed by an ARM core processor and a DSP core processor; inter-core communication is carried out between the ARM core processor and the DSP core processor through a shared memory and interruption; the ARM core processor is responsible for protocol analysis, task scheduling and man-machine interaction, and the DSP core processor focuses on signal processing and real-time control and integrates four communication systems, a channel evaluation module and the like. The problems that application of a single communication mode is limited and underwater acoustic communication links in different complex sea areas are unstable are effectively solved, and the deep sea area underwater vehicle information collaboration and return and deep sea area observation and monitoring network system capacity are effectively improved and guaranteed.

The invention relates to the technical field of distributed architectures, and discloses a multi-screen cooperative control method based on an OpenHarmony distributed architecture and an all-in-one machine system.The method comprises the steps that a broadcast frame found by equipment is sent through an OpenHarmony distributed soft bus, security authentication is conducted on a plurality of display terminals responding to the broadcast frame, a virtual bus channel is established, and the virtual bus channel is sent to the display terminals; obtaining a virtual device group; creating a distributed industrial data object on the main control equipment of the virtual equipment group and synchronizing the distributed industrial data object to each slave display terminal to obtain a distributed data model; mapping the anonymous shared memory to an annular buffer area; and writing real-time data acquired by the industrial field equipment into an annular buffer area, and transmitting the real-time data to each slave display terminal of the virtual equipment group, in the method, each display terminal accesses an agent instance of an industrial data object through a distributed data management framework; and the requirements of high reliability, low delay and high consistency on multi-screen cooperative control in an industrial field are met.

The invention discloses a data transmission optimization method and system in hybrid deployment of a real-time system and a non-real-time system based on Jailhouse, belongs to the technical field of data transmission, and solves the problems of low data transmission efficiency, high copy overhead, lack of universality of interfaces and the like in an existing system. Comprising the steps that a hybrid deployment platform including a real-time system and a non-real-time system is built on a multi-core processor platform, and the hybrid deployment platform runs in an isolation environment of a Jailhouse partition manager; building an abstract transmission interface layer as a unified channel for accessing a shared memory by a real-time system and a non-real-time system; establishing a plurality of shared memory channels in the abstract transmission interface layer and scheduling according to the priority of tasks; establishing a data transmission path, and compressing a memory copy operation into one time by adopting a memory mapping multiplexing technology and a DMA (Direct Memory Access) collaboration mechanism; and a cache consistency maintenance technology is adopted, so that the data consistency is ensured, and the optimization of data transmission is completed. The method is suitable for application scenes such as industrial automation, intelligent connected automobiles and edge calculation.

The invention discloses a real-time data receiving and transmitting system and method based on a kernel bypass, and the method comprises the steps: processing the received data when the data receiving and transmitting system receives the data, and achieving the high-level data with complete semantics; updating a write pointer through atomic operation, and writing high-level data into a lockless annular queue in the shared memory module; the upper-layer application module reads the lock-free annular queue in the shared memory; when the data receiving and transmitting system transmits data, upper-layer data to be transmitted is transmitted to the shared memory module, and a write pointer is updated; the upper layer data in the shared memory is read in a polling or interruption mode, a read pointer is updated, the upper layer semantic data is segmented according to the format, a UDP data packet is obtained, the UDP data packet is written into a network card in the polling or interruption mode, and data sending is achieved. Through the data receiving and transmitting system disclosed by the invention, microsecond-level ultra-low delay of data certainty is realized, and efficient decoupling and stable throughput are realized.

A method of VM migration includes allocating a designated system memory for a guest VM on a target node and reserving a shared physical memory region on a transfer node that is part of a shared memory pool that is memory coherent and cache coherent with a source node and the target node. The VM system memory of the guest VM is re-mapped from a first physical memory region on a source node to the shared physical memory region on the transfer node and the system memory of the guest VM is copied from the first physical memory region on the source node to the shared physical memory region on the transfer node. The designated system memory on the target node is mapped to the shared physical memory region on the transfer node; the guest VM is stopped on the source node and then resumed on the target node.

The invention provides a lightweight deployment method and system of a large model on edge computing equipment. According to the method, the acceleration capability portrait is constructed by extracting the hardware instruction set architecture type of the target edge device and the number of parallel computing units. Based on the instruction type, the large model weight is grouped, divided and pre-calculated through a unified lookup table vectorization engine, and a pre-calculation vector matched with the target instruction set is generated; and according to the number of the parallel units and the instruction-level parallel capability, compiling the pre-calculation vector to generate an adaptive parallel table look-up instruction block, distributing execution threads with the same number as the parallel units, and eliminating data dependence conflicts. And finally, loading the instruction block to a shared memory area, configuring topological logic of the photoconductive switch matrix based on an instruction type, and dynamically switching a data transmission path in a hardware instruction period. According to the method, the efficient deployment of the large model in the edge equipment and the low-delay reasoning in the resource-constrained environment are realized.

The invention provides an unmanned aerial vehicle calling system and method based on open source gap atomization service. The system comprises a gap equipment layer, an atomization service framework layer, a protocol hard mapping engine layer, a shared memory direct writing layer and a PX4 flight control layer which are in communication connection with one another. Wherein the gap equipment layer is used for carrying out atomization service standardization packaging on basic services of the unmanned aerial vehicle according to a gap system rule, and constructing an interaction channel between the unmanned aerial vehicle and various terminals; the atomization service framework layer is used for managing the atomization service of the unmanned aerial vehicle; the protocol hard mapping engine layer is used for constructing a hard coding mapping table of an unmanned aerial vehicle atomization service and a PX4 flight control instruction; the shared memory direct writing layer is used for constructing a direct connection framework between a shared memory area special for the unmanned aerial vehicle and a hardware module in a swan-gap system; and the PX4 flight control layer is used for carrying out direct execution on an instruction from the swan gap system through PX4 flight control and carrying out quadruple safety protection on flight control.

The hybrid addressing method and the hierarchical multi-core cascading device provided by the invention comprise the following steps: if a core in a basic module initiates a memory access request, analyzing the memory access request to obtain a row offset high order of an access address, if the row offset high order is located in a sequence region, rearranging the access address to obtain a rearranged address, and if the row offset high order is located in the sequence region, rearranging the rearranged address to obtain the hierarchical multi-core cascading device. The rearranged address is restrained in the memory bank of the basic module, the sequence area is a preset row in the memory bank of the core, and if a row offset high bit is located in an interleaving area, the access address is divided to obtain a divided address; wherein the divided addresses are stored in memory banks of different basic modules, and the interleaving area is an area except a sequence area. A dual-mode design of a local sequence region and a global interleaving region is adopted, and an exclusive sequence region is divided in global interleaving mapping by recoding an address high bit, so that balance between a global shared memory view and local low-latency access is ensured.

The invention discloses a data zero-copy processing method and system based on an eBPF zero-trust gateway, and belongs to the technical field of computer network security. According to the method, an XDP program is loaded on a network card driving layer, and an AFXDP technology is combined to construct a kernel mode and user mode shared memory pool UMEM; when an original message arrives, an eBPF unit mounted by an XDP program intercepts the message in real time, a tunnel packaging space is reserved in situ in an original memory, and if the head space is insufficient, a new memory page is logically mapped from a UMEM standby buffer pool to achieve extension. Identity fingerprints are generated by extracting message feature vectors, load in-situ encryption is completed in cooperation with a kernel kTLS module, and data are prevented from being returned to a user mode for processing. According to the method, multiple copying and context switching of a kernel-user mode of a traditional SSL VPN are eliminated, the throughput is improved by 2-3 times, the delay is reduced to the microsecond level, the CPU utilization rate is reduced by 40% or above, application layer DDoS attacks are effectively defended, and the method is suitable for 5G edge computing and cloud native zero-trust gateway scenes.

The invention discloses an embedded system resource allocation method and system based on hardware virtualization. The method comprises the following steps: deploying a Type-1 virtual machine monitor system on a multi-core processor hardware platform supporting virtualization extension; the Type-1 virtual machine monitor carries out virtualization, resource partitioning and security isolation on hardware resources, and creates operation system execution environment'domain 'partitions zoneG and zoneR. A universal operating system is deployed in a zoneG domain partition to take charge of a high-computing-power computing task, and a real-time operating system is deployed in a zoneR to take charge of a real-time computing task. Low-delay inter-core communication between a real-time operating system and a general operating system is realized based on an OpenAMP protocol, and data is transmitted through a shared memory and an interrupt mechanism; according to the requirements of real-time tasks and non-real-time tasks, a Type-1 virtual machine monitor dynamically adjusts a resource allocation strategy, and the priority and deadline of the real-time tasks are ensured.

The invention provides a deep learning communication operator data processing method and system and artificial intelligence computing equipment, and relates to the technical field of deep learning. The method comprises the following steps: under the condition that a processor core is started, distributing a target storage space for a temporary array of a plurality of threads in an on-chip shared memory; under the condition that a plurality of threads execute data processing in parallel, initializing a temporary array data area allocated by each thread in the target storage space; under the condition that a plurality of threads are processed in parallel and distributed to respective input data units, distributed calculation processing is carried out on the temporary array corresponding to each thread according to the input data units, and a temporary array calculation result corresponding to each thread is obtained; and performing reduction processing on the temporary array calculation result to obtain a global statistical result corresponding to the plurality of threads. According to the embodiment of the invention, by optimizing the storage management and parallel processing flow of the temporary array, the execution efficiency of a communication operator and the overall throughput of the system can be remarkably improved.