The BaaS Platform Orientated Research And Implementation For Scheduling Algorithm

With the concept of BitCoin put forward by SatoshiNakamoto in the year of 2009,the blockchain technology becomes more and more popular among Chinese industries.For a better development of blockchain,the community proposed the concept of Blockchain as a Service(BaaS)which runs the Blockchain Application on the Cloud Computing platform to provide storage service.The original scheduling problem and the affinity of private chain makes it a challenge for scheduling resource effectively.In my paper,we put more focus on resource scheduling specified for the consensus algorithm PBFT and blockchain,and put forward the static scheduling algorithm and self-adaption scheduling algorithm.(1)The static scheduling algorithm introduces resource balance analysis and affinity analysis to ensure a balanced resource allocation,high service availability and physical disaster recovery.The innovation of each process can be briefly described as follows:First,the input data preprocessing is performed before the algorithm's execution,that is,the input parameters of the algorithm are estimated according to the characteristics of the blockchain itself;Secondly,the Best-Balanced algorithm considers the balance of resource allocation from three perspectives,including Load Balance(LB),Remaining Resources(RRB)and Resource's Balance(RB);Thirdly,an affinity analysis and reordering algorithm is proposed to solve the problem of affinity among blockchain applications according to the characteristics of private blockchain.(2)The self-adaption scheduling algorithm solves the problem of service collapse caused by the dynamic change of resource usage in cloud computing platform using early warning and relocation.The innovation points can be summarized as follows:The prediction model in the early warning is based on ARIMA time series,and introduces the quantitative correction of the prediction results;The migration screening uses the Filtering Based On Priority(FBP)algorithm to ensure the safety of migration.Finally,we do the system experimental analysis and the algorithm simulation analysis.The experimental results show that the static scheduling algorithm and the adaptive scheduling algorithm can effectively solve the balance scheduling,affinity scheduling and dynamic scheduling problems of the blockchain application on the cloud platform,and the system experimental analysis further verifies the practicability of the algorithm.