Design And Implementation Of Task Scheduling Module Of Could Security Vulnerability Scanning System

Based on the statistical data, most of the network security issues are caused by found vulnerabilities. Vulnerability scanning technique is significant to protect user’s network by scanning the machines automatically. However, current vulnerability scanning systems are all stand-alone oriented, which leads to heavy computing cost and unsatisfied to large scale scanning request. The development of cloud computing offers us a new solution, SECaaS. With the help of distributed technology, we can offer security vulnerability scanning service with the extensible resources on cloud. This solution not only. reduces the computing cost, but also lowers the training requirements for users. Task scheduling is the key part of security scanning system deployment on cloud platform. The schedule algorithm should take users’different QoS requirements into consideration.The applications of public cloud have large amount of users, data and needs. In this case, to schedule the scanning requires efficiently and make good use of the extensible dynamic resources is the emphasis and difficulty to achieve the aim of high throughput capacity and good user experience with low cost.This thesis designs a mechanism to schedule the scanning tasks based on prediction task data and real-time resource workload. To predict the coming scanning tasks, this thesis uses the network traffic prediction model. And to schedule the tasks, we define a valuation factor evaluating the scheduling strategy. This schedule algorithm can support differentiation service, with the throughput capacity guaranteed. And the deployment and testing, including unit tests and deployment tests, are also done to guarantee its availability, reliability and performance.