2970 results about "Pool" patented technology

Systems and methods are described for a implementing a streaming data processing system that includes a pool of pre-configured resources and a pool of dedicated resources. The streaming data processing system can implement a processing pipeline using compute resources. The pool of pre-configured resources can support previews of processing pipelines for a plurality of users and the pool of dedicated resources can support full deployments of processing pipelines for a particular user. The streaming data processing system can implement a preview of a processing pipeline using a pre-configured resource of the pool of pre-configured resources. Further, the streaming data processing system can implement the processing pipeline using a dedicated resource of the pool of dedicated resources. The streaming data processing system can provision the dedicated resource and deploy the processing pipeline using the dedicated resource.

A disclosed method facilitates translation of natural language queries into query language statements usable to retrieve data from or write data to a particular database. The method includes obtaining a pool of shots. Each shot in the pool includes a natural language query component and a corresponding database translation component. The method further provides for vectorizing the natural language query component for each of the shots into a common vector space; receiving a natural language query from a user interface; vectorizing the natural language query within the common vector space; identifying a subset of vectorized natural language query components that satisfy a similarity metric when compared to the vectorized natural language query; and generating an LLM prompt that includes shots from the pool corresponding to the subset of the vectorized natural language query.

The invention provides a security encryption communication method and system based on quantum key management. The method comprises the following steps: acquiring system state data, and generating node state data, network connection diagram data and a system state matrix; receiving a communication request data packet, and generating a request feature vector, a service weight coefficient and a service demand matrix; the method comprises the following steps: collecting real-time network state data, constructing an enhanced Bayesian network model, and generating a network state scoring matrix; acquiring key pool state data, establishing a key consumption prediction model, and generating a key resource distribution matrix; constructing a multi-objective optimization model, and generating a scheduling strategy matrix; and executing the scheduling strategy and performing performance evaluation. According to the method, the quantum state de-coherence effect and the channel noise influence are considered, efficient management and safe distribution of quantum keys are achieved by establishing a complete evaluation and prediction system, and the performance and reliability of a communication system are improved.

The invention discloses a real-time interactive digital human system supporting high concurrency and an implementation method thereof, and relates to the technical field of digital human interaction.The system comprises a model instance pool module used for loading the weight of a deep learning model to a shared memory area through the memory mapping technology; the multi-thread scheduling module is used for scheduling user requests by adopting a lock-free queue and a dynamic priority algorithm; the asynchronous pipeline processing module is composed of decoupled micro-services, and the modules are connected in series through asynchronous message queues; the client SDK is used for dynamically switching a rendering mode according to terminal hardware performance and network conditions; the elastic capacity expansion and contraction module is used for monitoring resource loads in real time and automatically adjusting the number of service instances; and the audio and video synchronization calibration module is used for ensuring that the synchronization error of the audio and video frames is lower than a preset threshold value through a timestamp alignment algorithm. According to the scheme, core challenges in a high-concurrency digital human interaction scene can be systematically solved, and innovative support is provided for large-scale real-time application.

The invention discloses a code generation method and device based on LLM multi-agent cooperation and computer equipment. The method comprises the following steps: acquiring a user programming demand; analyzing the user programming demand based on LLM and decomposing a programming task to generate a task tree; selecting a to-be-allocated sub-task from the task tree, and evaluating the matching degree of an intelligent agent in an intelligent agent pool and the to-be-allocated sub-task based on LLM to generate a task allocation instruction; sending the task allocation instruction to a corresponding target agent, so that the target agent executes the to-be-allocated sub-task, and feeding back an execution result; and outputting the execution result. By implementing the method provided by the invention, the programming efficiency and quality can be improved, the resource allocation can be optimized, and the efficient development of complex requirements and large projects can be supported.

A database of text associated with different domains is maintained. Large language models (LLMs) are prepared for use in the different domains by providing the associated text to an instance of an LLM. Thus, using multiple instances of the same pre-trained LLM, domain-specific LLMs are generated. The text provided to the LLM instance may be selected based on an account identifier of the user accessing the LLM, the tenant accessing the LLM, a user selection of a domain, or any suitable combination thereof. A pool of prepared LLM instances may be generated before the access request is received. If a response provided by an LLM instance in a domain to a prompt was rejected by a user and additional information was received during the session to improve the response of the LLM instance, the additional information may be to the text used to prepare future LLM instances for the domain.

The invention relates to the technical field of data storage, can be applied to business scenes such as medical health and financial science and technology, and discloses a multi-Pool storage resource adaptive writing method which comprises the following steps: monitoring available storage space, read-write input / output rate, average response time and data access frequency of each Pool; dynamically adjusting the current write-in weight of the Pool according to the monitoring result; converting the current write-in weights of all Pool into probability values, and constructing a discrete probability distribution table; selecting a target Pool through a pseudo-random number generator according to the probability distribution table; and writing the to-be-stored data into the target Pool. According to the method, the running state of each Pool is continuously monitored, and the writing weight is adjusted in real time, so that the problem of resource allocation rigidity under a fixed weight strategy is avoided; in combination with a discrete probability distribution table constructed based on normalized weight and a pseudo-random selection mechanism, selection of the target Pool has certain load balancing elasticity, and data can be more effectively guided to be written into the Pool with light load and excellent performance.

The invention discloses an intelligent private cloud resource scheduling management system based on a hyper-converged architecture, particularly relates to the technical field of virtual resource scheduling management, and is used for solving the problem of overall utilization rate reduction caused by cascade resource oscillation and fragmentation due to fault recovery in the prior art. Candidate buffer nodes are screened based on the resource occupancy state, a migration priority queue is generated in combination with the node resource isomerism matching degree and the real-time load, and an independent buffer resource pool is reserved at a target node; key resources are directionally reserved through resource locking labels, task dependence periods are associated, and task conflict risks possibly triggered after the resources are released are dynamically predicted; adjusting a label effective period to cover a task execution window based on a conflict identification result; and removing the label in the preset time window, releasing the resources to the buffer pool, synchronously triggering global resource integration and candidate node list updating, and realizing efficient recovery and dynamic redistribution of fragmented resources, thereby inhibiting multi-node resource fluctuation.

A system can receive, from a remote computer of remote computers, inventory data representative of an inventory, wherein the inventory data indicates that the remote computer is configured to implement a shared memory pool with the remote computers according to a compute express link protocol. The system can receive, from the remote computer, a request to onboard the remote computer, wherein the request adheres to a defined security protocol and data model architecture, and wherein the system is configured to remotely manage the remote computers as part of a hybrid cloud platform that comprises the remote computers. The system can authenticate the remote computer based on the request and according to the defined security protocol and data model architecture. The system can implement the shared memory pool for the remote computers.

In at least one embodiment, processing can include: receiving write operations; persistently recording, in a write cache or log, the write operations using page descriptors (PDESCs) of a PDESC pool and page buffers (PBs) of a PB pool; selecting, in accordance with criteria, write data pages stored in the PB pool for demotion to a physical large block (PLB) pool included in backend non-volatile storage, wherein each write data page selected denotes content written by a corresponding one of the write operations; responsive to the selecting, persistently storing the write data pages of the PB pool in the PLB pool; and updating PDESCs associated with the write data pages to reference corresponding storage locations in the PLB pool rather than in the PB pool.

The invention discloses a compatible operation method for an Android camera HAL in a container based on memory access virtualization, and the method comprises the steps: taking a DMA-Buf memory heap of a Linux kernel host system as a target memory heap, taking an ION memory heap of an Android container system as a source memory heap, building a process exclusive memory management context, an FD cache table and a synchronous fence pool, and carrying out the execution of the process exclusive memory management context, the FD cache table and the synchronous fence pool; kernel registration and node binding of virtual ION equipment, pre-allocation of a first memory pool and access hook registration of kernel layer equipment are completed, and when an ION file descriptor is obtained in an HAL process, legality of a container process is verified, and context binding is initialized to the file descriptor; intercepting a memory allocation request of the HAL process, analyzing and adapting parameters, and preferentially multiplexing first memory pool resources to obtain an ION handle; when a data sharing request is processed, matching an FD cache or generating a new FD through an ION handle; when the HAL process releases the memory, resources are recycled according to a memory source, and when the HAL process exits, the context is cached or destroyed, so that cross-system memory operation compatibility is realized.

The invention discloses a processing method for improving batch inference efficiency of a large language model, which comprises the following steps of: 1) executing pre-filling processing on all inference tasks to be processed by utilizing the large language model to obtain a key vector and a value vector of each lexical element in a sequence of each inference task and a lexical element, and caching the key vector and the value vector and the lexical element into a task pool; 2) during the first round of iterative calculation, forming a tensor TB by using the latest single lexical elements T corresponding to n reasoning tasks selected from a task pool, forming a tensor KVB by using corresponding key value vectors, and forming a tensor AB by using attention mask vectors; inputting the TB, the KVB and the AB into a large language model for reasoning, calculating and updating the TB, the AB and the KVB for the next round of iterative calculation; and 3) after the iteration of a reasoning task is finished, splicing the lexical elements output by the reasoning task in each iteration to obtain a reasoning result of the reasoning task, and selecting a to-be-processed task from the task pool to execute the next iteration. The task processing efficiency and the computing power resource utilization rate are improved.

The invention relates to the technical field of page rendering, and discloses a progressive page rendering method and system based on streaming data, and the method comprises the steps: carrying out the data segmentation through obtaining a data stream, user interaction data and equipment performance parameters, and obtaining data blocks; then, performing cache management in combination with the data, and constructing a multi-level cache pool; thirdly, performing priority grading and sorting on the data blocks to form a rendering task queue; and according to the equipment performance parameters, optimizing a task sequence and obtaining an optimized task sequence. Next, combining the optimized task sequence and user interaction data, predicting data about to enter a viewport, and generating a viewport pre-rendering task; and finally, according to the viewport pre-rendering task, the multi-level cache pool and the equipment performance parameters, performing rendering strategy optimization to obtain a dynamically adjusted rendering task flow. The method can realize dynamic resource scheduling.

Systems, methods, and data storage devices for host storage command management for dynamically allocated floating namespaces are described. A data storage device may support multiple host namespaces allocated in its non-volatile storage medium and include a floating namespace pool that includes at least some data units from those host namespaces. Host storage commands to be processed using the floating namespace pool may be received and payload sizes may be determined. A next host storage command may be determined based on the relative payload sizes and executed using a data unit from the floating namespace pool, for example, based on allocating virtual command queues to the floating namespace pool and sorting the incoming host storage commands by payload size.

The invention belongs to the technical field of computing power scheduling, and particularly relates to a computing power distribution scheduling method and system for an edge computing platform, and the method comprises the steps: generating a business pool based on a historical resource request task; constructing a business computing power demand prediction model based on XGBoost, adding the influence factors into the prediction model, generating resource request business prediction output of each node device, updating the business pool according to the prediction output, evaluating the computing power scale required by the resource request business in the updated business pool, and obtaining the computing power demand; according to the computing power demand obtained by each node device, taking energy consumption minimization and response time minimization as targets, carrying out multi-target solution based on an improved firefly algorithm, and obtaining a computing power distribution scheme which enables the energy consumption and the response time to reach the optimal; and allocating schedulable computing power resources in the edge computing platform based on the computing power allocation scheme. According to the method, overload of part of the data centers can be avoided, meanwhile, the overall calculation efficiency is improved, and calculation power resources are reasonably utilized.

The invention relates to the technical field of message queues, in particular to a message controllable optimization method and system based on kafka, and the method comprises the steps: obtaining a data attribute, and calculating a data priority value according to the data attribute; the method comprises the steps of establishing a position index cache mechanism based on priority data, persisting partition-offset data in a cache pool to an index table according to the position index cache mechanism, and performing priority consumption of high-priority messages based on persistent priority indexes, including setting a consumption control program architecture, performing priority consumption on the high-priority messages, and performing priority consumption on the high-priority messages based on the persistent priority indexes. Isolating the consumption queue and introducing a message de-duplication operation; according to the method, full-link performance adjustment and optimization are carried out on the premise of preferential consumption of high-priority messages, service scene extension is carried out based on the adjusted and optimized full links, the messages are pulled directionally according to the priorities through a consumption control program, the disorder of parallel consumption of Kafka partitions is broken through, the high-priority messages can directly jump to corresponding positions to be processed preferentially, and the service scene extension efficiency is improved. The consumption delay is obviously shortened.

The invention discloses an operation management method and system for a server cluster, and belongs to the technical field of server operation management, and the method comprises the steps: constructing a computing power unit pool containing a plurality of service blocks, combining the resource state of a function module with application label information, recognizing the operation situation of the function module, and generating a scheduling adaptation factor model, link capacity, position distribution and communication delay among service blocks are further fused to construct an expansion structure map, a deep self-adjusting engine is adopted to carry out offline reinforcement learning training, and an optimal capacity expansion strategy under multiple situations is generated; the system can dynamically judge the capacity expansion behavior based on the current state, and performs strategy updating by continuously collecting capacity expansion feedback, so that intelligent, self-adaptive and high-efficiency control of resource scheduling is realized.

In some implementations, the system may include one or more processors, coupled to memory, the processor configured to set in the memory a pool of shared send queues (SSQs), each SSQ in the pool of SSQs is configured for use as a send queue (SQ) for one or more queue pairs (QPs). The one or more processors may allocate at least one of SSQs from the pool of SSQs to the QP, the SSQ set for a process having a plurality of connections to a plurality of remote processes. The one or more processors may send by the SSQ, via the plurality of connections, outgoing messages to separate remote processes of the plurality of remote processes. Also, the system may include implementations where the pool is set responsive to a first number of the QPs reaching a threshold. The system may include implementations where the pool of SSQs is pinned in the memory.

The invention discloses a computing system, a computing method, computing equipment, a medium and a program product in the technical field of computers. The heterogeneous acceleration card with the built-in Ethernet controller forms the heterogeneous acceleration pool, the heterogeneous acceleration pool is connected with the Ethernet switching module in an Ethernet mode or a remote direct data access mode, and the Ethernet switching module is connected with the computing resource pool; the management core in the computing resource pool can manage the acceleration cards in the at least one heterogeneous acceleration pool and the binding relationship between the computing core and the acceleration cards, and distribute the processing task to the acceleration cards in the at least one heterogeneous acceleration pool, so that the acceleration cards in the at least one heterogeneous acceleration pool cooperatively complete the processing task. Wherein data transmission is accelerated in an Ethernet mode or a remote direct data access mode, so that not only can sufficient computing power support be provided for task operation, but also high bandwidth and low delay can be provided for high-speed communication and large-data-volume transmission of tasks, and rapid and stable operation of model training tasks is facilitated.

The invention provides a memory pooling method and system for model reasoning acceleration and a computer program product, and relates to the technical field of computers. The model is a separation pooling architecture, the separation pooling architecture comprises a pre-filling node pool, a decoding node pool and a CXL memory pool which are separated, and the memory pooling method for model reasoning acceleration comprises the steps that a reasoning request is allocated to a first pre-filling node in the pre-filling node pool based on a first scheduling strategy; the first pre-filling node processes the reasoning request to obtain a key value cache; storing the key value cache in a CXL memory pool; selecting a first decoding node in the decoding node pool based on a second scheduling strategy; the first decoding node obtains a key value cache from a CXL memory pool based on the reasoning request; and the first decoding node generates a reasoning result according to the obtained key value cache.

The invention relates to a task scheduling method and device, computer equipment, a storage medium and a program product. The method comprises the steps of determining priorities of tasks to be executed in at least two task queues according to a preset time period, and updating the at least two task queues according to the priorities; according to the updated task sequence of the at least two task queues, the resource demand of each to-be-executed task and the resource specification of each node in the node pool corresponding to the at least two task queues, allocating each to-be-executed task to a matched node for processing; wherein the type of the node pool comprises a reasoning type and a training type, and each task queue corresponds to one type of node pool. By adopting the method, the priority of the to-be-executed tasks in the task queue can be periodically updated, the situation that the to-be-executed tasks with different priorities cannot be scheduled is avoided, meanwhile, the to-be-executed tasks are allocated to more appropriate nodes to be processed, the task processing efficiency can be improved, and the task fine scheduling effect can be optimized.

This application provides a semi-persistent scheduling method, a user equipment, a base station, and a communication system. In the method, a user equipment receives first configuration information and second configuration information that are sent by a base station. The first configuration information includes an SPS resource configuration pool including one or more pieces of SPS resource configuration information and an index number corresponding to each piece of SPS resource configuration information. The second configuration information indicates at least one index number in the SPS resource configuration pool and a sequence of the at least one index number.

This specification is generally directed to techniques for generating interfacing source code between computing components based on natural language input. In various implementations, a natural language input that requests generation of interfacing source code to logically couple a first computing component with a second computing component may be processed to generate an interface request semantic embedding. The interface request semantic embedding may be processed based on one or more domain models associated with the first and second computing components to generate a pool(s) of candidate code snippets for logically coupling with first and second computing components. A plurality of candidate instances of interfacing source code may be generated between the first and second computing components. Each candidate software interface may include a different permutation of candidate code snippets from the pool(s) of candidate code snippets.

The invention discloses a park management method based on digital twinning, particularly relates to the field of park operation, and is used for solving the problem that a multi-source event is difficult to stably chain under the conditions of compensation delay and evidence gap and causes misalignment of disposal triggering. Comprising the following steps of: registering an event in a pool to generate an event number and writing the event number into a platform arrival moment; screening and mapping an anchor point event to a park object identifier and a process stage identifier to generate a management token; checking a stage sequence and an evidence item requirement set according to an event chain rule table in event chain construction and generating a conflict point list; the delay portrait library generates a credible mark when a source is generated and forms a reverse sequence violation spectrum; the digital twin object relation graph forms an influence domain span mark; the closed table is returned to obtain a disposal convergence guarantee level; the cost-sensitive decision tree outputs a disposal template identifier, and the disposal template table generates a management action instruction and a trigger voucher and records an instruction acceptance identifier; and returning, comparing and freezing the exception management token, and updating the event chain rule table and the delay portrait library.

The invention belongs to the technical field of distributed data access, discloses an RDMA (Remote Direct Memory Access) distributed data transmission method supporting high-concurrency message transmission, and aims to improve system performance in a high-concurrency message transmission environment. The method comprises the following core steps: 1) peer-to-peer server design: each device is an independent server, and connection is established by a connection channel monitored at any time; 2) establishing a connection resource pool in advance; 3) creating and registering an RPC message; 4) transmitting the RDMA to a memory mailbox pool of a target server; 5) the client and the server perform high-concurrency processing; 6) dynamically polling the server; (7) zero-copy transmission of data is achieved through the RDMA technology. According to the invention, the dynamic expansion capability of the architecture is improved by adopting a connection resource decoupling mode, QP jitter can be reduced and the stability of the system can be enhanced by registering a memory of a continuous large page at one time, and the message transmission efficiency under the distributed network architecture is effectively improved by adopting a combined sending strategy.

An adapter to a base model of an artificial intelligence (Al) system is disclosed. The adapter includes a connector to connect the adapter to the base model such that during an operation of the Al system at least some portion of data transformed by the base model is propagated from the base model to the adapter and back from the adapter to the base model. The adapter includes a non-linear modifier to modify the data received from the base model non-linearly before returning the modified portion of the data back to the base model, and an Al trainer to tune the non-linear modifier of the adapter by propagating training data through the base model and the adapter and updating weights of the non-linear modifier of the adapter for given weights of the base model to optimize a loss function. Further, weight matrices for the base model and the adapter are jointly constructed by an additional module, which efficiently uses a pool of parameters to allocate to save memory requirement for adaptation of the Al system.

The invention discloses a Direct3D rendering model compatible method based on a dynamic template pool, which comprises the following steps: establishing three types of mapping tables between D3D and Vulkan when compiling DXVK, constructing resource metadata, and creating a core, extended and temporary three-level template pool according to the mapping tables after starting; when the D3D application creates resources, metadata is initialized, parameters are verified, physical memories are allocated and grouped, resource groups are pre-verified, a batch binding command is generated, memory binding is completed, and the metadata is updated; when a resource view is created, a template is matched from a template pool, a handle is generated after instantiation, a resource handle is associated, and a descriptor is bound; when a rendering state is set and a rendering instruction is executed, the rendering state and the rendering instruction are respectively converted into a Vulkan related state and a Vulkan related instruction through a mapping table, a handle is bound after a PSO cache is inquired, a command buffer area is submitted to a GPU queue to execute drawing, and compatible operation of the D3D application on a platform supporting a Vulkan operating system is realized under the condition that the GPU does not support VKKHRmaintence5 and VKKHRmaintence6 extension.

The invention discloses a file transmission protection method, system and device based on multiple encryption algorithms, and relates to the technical field of information, and the method comprises the steps: selecting a target file needing to be transmitted, carrying out the preprocessing, and generating file fragments; analyzing the sensitivity of each file fragment, and performing hierarchical encryption processing to obtain an encrypted fragment set; selecting a steganography carrier file, and embedding the encrypted fragments into the carrier file to generate a steganography index table; generating a dynamic session key based on the confusion key pool, encrypting the steganographic file transmission layer and embedding a dynamic key generation parameter; the receiving end generates a session key according to the transmission message header, decrypts the transmission layer data, extracts the steganographic file and decrypts the steganographic file; recombining the decrypted and recovered fragmented data into a complete file according to an identifier sequence; by dynamically matching the steganography carrier file, the efficient embedding of the fragmented data and the generation of the steganography index table are realized, and the concealment and the adaptability in the transmission process are ensured.

The present application belongs to the technical field of data processing and relates to a network message processing method and apparatus, and a computer device and a storage medium. The method comprises: on the basis of a received target network message, acquiring a target memory block from a pre-constructed shared memory pool, so as to generate a message memory; on the basis of a network-interface-card driver, performing first identification processing on the target network message, so as to obtain a target network message descriptor; on the basis of the target network message descriptor, performing second identification processing on the message memory, so as to obtain a first socket buffer; and performing parsing processing on the first socket buffer by means of a network protocol stack, and on the basis of a parsing processing result, reading network message data from a virtual memory corresponding to the message memory, so as to complete the processing of the network message. In the present application, on the basis of a constructed shared memory pool, zero-copy network packet reception can be realized during network message processing, thereby avoiding performance loss caused by the memory copy of messages.

The invention relates to the technical field of artificial intelligence and natural language processing, and discloses a task processing method and device, computer equipment and a storage medium, and the method comprises the steps: carrying out the dynamic batch processing of an input text task, and obtaining a conventional batch and an ultra-long batch; detecting a task belonging to an ultra-long batch in the to-be-processed text task as a priority text task, and realizing preemptive scheduling of the current text task by the priority text task through a hardware interrupt mechanism; calculating the priority text task by adopting a local-global mixed attention mechanism; calculating the text tasks belonging to the conventional batch by adopting a first sparse rate, and generating a first historical key value pair; calculating a priority text task by adopting a second sparse rate, and generating a second historical key value pair; and caching the first historical key value pair in a first cache pool, and caching the second historical key value pair in a second cache pool. The method can be applied to the field of financial science and technology businesses, and efficient differentiation processing and computing resource optimization of super-long and conventional text tasks are achieved.