Task scheduling algorithm for cost perception under cloud environment

Abstract

The invention discloses a task scheduling algorithm for cost perception under cloud environment, relates to the application field of distributed data management, cloud computing and the like, and provides a universal task scheduling method. Both interactive tasks and analysis tasks can be processed, and it is ensured that the cost is the minimum in the task processing process. The method comprisesthe steps of receiving all tasks submitted by a user, and adding the tasks into a task queue; combining a waiting queue to the task queue, and emptying the waiting queue; according to the task type,calculating resources needed by the tasks in the task queue; according to the requirement for the resources, ranking all the tasks in the task queue from more to less; scheduling the ranked tasks oneby one. The unified scheduling algorithm is provided for scheduling the interactive tasks and the general analysis tasks, and the user's satisfaction and task processing throughput are improved.

Description

technical field [0001] The present invention relates to a cost-aware task scheduling algorithm in a cloud environment, which involves application fields such as distributed data management and cloud computing, and proposes a general task scheduling method, which can handle both interactive tasks and analytical tasks. In the process, the minimum cost is guaranteed at the same time. Background technique [0002] Big data and cloud computing have become the new darlings of today's IT industry. Major IT companies have launched commercial cloud platforms and big data processing solutions. Many research institutions have carried out theoretical research and application exploration on the core technologies of cloud computing and big data. Governments of various countries have also included cloud computing and big data in their national plans and invested heavily. At present, cloud computing has been applied to many fields, and various open source platforms and commercial cloud ser...

AI Technical Summary

Problems solved by technology

However, interactive tasks and analysis tasks require the same resources and there is no relationship between tasks, but the optimization goals of the two types of tasks are different, so there are currently few scheduling methods that can handle interactive tasks and analysis tasks at the same time

[0005] To sum up, although the current task scheduling methods in the cloud environment mainly focus on task processing time as the optimization goal, the measurement of user satisfaction is not clear, and different scheduling algorithms are designed for interactive tasks and analysis tasks. The introduction of concepts such as task priority increases the complexity of the system, and at the same time does not significantly optimize the cost of task scheduling

Images