Research And Implementation Of Task Management System In Security Operation Center

With the rapid development of computers,communications,and virtualization technologies,and with the rapid commercialization of 5G,network cloud computing has become the preferred platform for more and more enterprises and individual users to perform their own series of personalized tasks.The security operation center platform is also a cloud computing platform.Users pay to choose to submit website asset scanning tasks on the security operation platform at the corresponding time period.The operation center will schedule the tasks submitted by users to the corresponding virtual machines,occupying a certain amount of time.Bandwidth of hardware and network resources to perform scanning tasks for the user's unique assets.However,the business of secure cloud computing platforms also has characteristics that are different from traditional cloud computing platforms.First,the security task's scanning period is relatively loose,with low requirements for real-time execution,high accuracy requirements,and even users sometimes require that security tasks be scheduled to be executed in the early morning idle periods.Second,the security task execution period and task content are relatively fixed,and the scale is medium.It is usually repeated on a daily or weekly basis.The users are relatively fixed,but they have different payment levels and belong to the toB business.How to ensure the QoS experience of different users is the key.Third,due to the limited hardware conditions,the load pressure of the security operation center is relatively large,and a good task balancing algorithm is needed to support it.At the same time,it is also necessary to ensure that the system can complete as many tasks as possible in the same time to improve overall revenue.Due to the many characteristics of the security operations center,in this paper we innovatively introduce an adaptive task scheduling system driven by historical data.By collecting the historical data of the system,we can automatically classify the complexity of the task by clustering based on the recent performance of the task.Then,we optimized the scheduling algorithm from two dimensions of time and space,and proposed two operating modes for users to choose,one is immediate execution and the other is intelligent time allocation.We have designed the heuristic task assignment algorithm and priority rules in detail to improve the QoS experience for users at different levels.The experimental simulation results show that this algorithm is more suitable for the business requirements of the security operation center,and has better load balance,resource utilization and system stability.