A method and system for intelligent scheduling of medical data computing tasks

Abstract

The invention discloses a method and system for intelligent scheduling of medical data calculation tasks. Firstly, a task queue is constructed. The task queue is divided into a sorting task queue and a queue waiting queue connected end to end. The sorting task queue eliminates unfinished tasks within a specified time. Enter the waiting queue, the head element of the waiting queue enters the sorting task queue, calculate the prediction model according to the task data in the queue, use the prediction model to predict the execution time of the new task, and calculate the time consumption in the queue according to the difference between the predicted time consumption and the actual time consumption The task priority is adjusted, and whether to update the model is judged according to the determination coefficient of the model. In the case of dynamically adjusting the execution of scheduling tasks, the present invention can timely adjust tasks that occupy computing resources for a long time, so as to prevent hardware resources such as servers from being in a high-load state due to a single resource, and make full use of resources after task adjustment is suspended. , reducing the waiting time for tasks to execute, and improving the efficiency of overall task execution.

Description

technical field [0001] The invention belongs to the technical field of medical data calculation, and in particular relates to an intelligent scheduling method and system for medical data calculation tasks. Background technique [0002] In the field of medical computing services, the main function of the multi-center intelligent medical information technology platform is to provide a scientific research platform for computing services for medical researchers. Researchers or medical workers can submit computing tasks of different dimensions in different hospitals. The servers in the hospital provide data support and computing services. At the same time, different hospitals can uniformly submit specific dimension computing tasks to the corresponding hospital services, realizing the display of computing service results across hospitals and laying the foundation for further scientific research. With the in-depth development of scientific research, the current demand for medical r...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to aim at the computing scenarios in the existing medical computing service field, aim at its shortcomings, design and invent a computing task intelligent scheduling method and system using a special scheduling algorithm based on double-queue prediction execution time, the present invention through the past in The different task records submitted by different business scenarios mainly include the performance information of the physical machine during execution, and the information of the submitted tasks (including the scale of processing data, the number of split tasks, task scenarios, etc.), In this way, the relationship between tasks and performance and task execution time can be obtained. When a new task is submitted, the duration of task execution can be predicted based on the existing information, so as to re-determine the order of task execution scheduling to obtain the number of tasks in equal time intervals. Executed the most, making full use of computer performance and reducing the occurrence of starvation

Images