5results about How to "Improve system resource utilization" patented technology

The application provides a terminal and edge server cooperative model reasoning optimization method and system and a storage medium. The method comprises the following steps: acquiring the computing power parameters and communication parameters of local terminals and edge servers in an initial model deployment environment; combining the computing power parameters, the communication parameters and layer structure information of the initial model to calculate a total splitting cost; taking the total splitting cost and model reasoning accuracy as optimization objectives, processing the initial model based on an optimal pruning strategy to generate a lightweight model; selecting a target splitting point matched with a real-time communication rate from the matching relationship between the communication rate interval and the splitting point, splitting the lightweight model into a terminal-side sub-model and an edge-side sub-model based on the target splitting point, deploying the terminal-side sub-model on the local terminal and deploying the edge-side sub-model on the edge server to carry out cooperative reasoning. The embodiment of the application can ensure reasoning accuracy while realizing low delay.

The application provides a financial product index calculation method and device and related products, and relates to the technical field of data processing. The method comprises the following steps: receiving a financial product index calculation request; if the request involves at least two financial products, the request is packaged as a parent task and sent to a first message queue; a first type of calculation node obtains the parent task from the first message queue, splits the parent task into at least two subtasks, and sends the subtasks to a second message queue; a second type of calculation node obtains the subtasks from the second message queue, and performs index calculation based on index calculation parameters; the first type of calculation node monitors a result storage layer, and when the calculation results of all associated subtasks are stored, the first type of calculation node performs summary processing to obtain a summary result and stores the summary result; and the summary result is returned to a user end. According to the application, physical isolation and parallel processing of parent tasks and subtasks are realized through double message queues, task blocking and deadlock risks are effectively avoided, and the efficiency of large-batch financial product index calculation and system resource utilization are significantly improved.

The invention relates to the technical field of data processing, in particular to an exhibition and display interaction logic processing method and system and a storage medium. Comprising the steps of collecting a user interaction request and a historical access log to form an initial execution path set; constructing a branch model, and identifying a high-frequency condition judgment sequence; if the sequence is not matched with the operation data, optimizing to obtain an adjusted condition judgment rule; extracting real-time features of the user interaction request from the operation data to generate feature vector representation, and traversing the adjusted rule to determine an optimal execution path; comparing the optimal path with the initial path set to obtain a path optimization index and generate a resource allocation adjustment scheme; when the response delay is reduced, the condition judgment rule is updated to form a final interaction processing mechanism; follow-up interaction requests are processed accordingly, and the execution efficiency and the dynamic adaptation capacity are evaluated. The problems that an existing exhibition and display system is fixed in interaction logic and difficult to adaptively optimize are solved, and dynamic optimization and efficient response of exhibition and display interaction logic are achieved.

The invention discloses a work scheduling method for a deep neural network reasoning task. The work scheduling method is applied to a vehicle edge computing system comprising a plurality of roadside units. The method comprises the following steps: firstly, constructing a system model, and modeling an inference task into a directed acyclic graph to distinguish executed and unexecuted layer nodes; establishing a communication time delay and energy consumption model, and defining a system total cost function as a weighted sum of task completion total time delay and intermediate result transmission energy consumption; modeling a scheduling decision process into a multi-agent Markov decision process, and taking each roadside unit as an independent agent; solving by adopting a multi-agent depth deterministic strategy gradient algorithm to generate an optimal scheduling strategy; and finally, determining a target roadside unit according to a strategy, and transmitting an intermediate result for relay calculation. According to the invention, the problems of service interruption caused by high mobility of the vehicle and lack of cooperation of multiple roadside units are effectively solved.

Embodiments of the present invention relate to the field of vehicle to vehicle communications technologies, and provide a resource energy determining method and apparatus, to resolve prior-art problems that system resource utilization is reduced, and that a resource selection collision easily occurs. The method includes: obtaining, by first UE, time-frequency resource indication information and resource cycle reservation information of second UE in a sensing time window of the first UE; determining, by the first UE, energy of a first resource based on the time-frequency resource indication information of the second UE; and determining, by the first UE, energy of at least one second resource based on the energy of the first resource and the resource cycle reservation information of the second UE, where a time-frequency domain location of the first resource in a transmission period in which the first resource is located is the same as a time-frequency domain location of each of the at least one second resource in a transmission period in which the second resource is located.