Development Of A Private Cloud Based On OpenStack-based Astronomical Data Processing

Astronomical data processing is very important part of astronomical research.With the rapid growth of function and observation capability of the new generation of telescopes,build a high performance real-time computing platform in observing sites and handle the data quickly have become a trend.OpenStack is an efficient technology of open source cloud platform,it has the features of abundant functions and it can meet the needs of astronomical data processing.Based on the application of OpenStack cloud platform technology in the astronomical data processing area,this thesis analyses the architecture of OpenStack and implemented an OpenStack private cloud of astronomical data processing.This thesis has completed the following work:First,through the research of OpenStack architecture,this thesis realizes the deployment-automation of OpenStack private cloud,classified into control node and compute node.This deployment can effectively reduce the time to build a private cloud and save manpower resource consumption;Second,the thesis design and realize an automatic management of private cloud computing resources,the model can automatically increase or decrease the computing resources based on the current computation of cluster in private cloud computing.Experiments show that this pattern can meet the needs of astronomical data processing completely and it is more efficient than other traditional data processing method of static computing resources allocation;Third,developed the function of automatic generation of UVFITS files for private cloud with image technology;developed the function of astronomical observation data storage with Swift component which satisfies the usual practice of researchers.Overall,the thesis combines OpenStack cloud platform technology with astronomical data processing,and implemented the private cloud of astronomical data processing.It can efficiently save our energy bills and reduce the observing costs.This pattern have certain value for reference to build a high-performance computing platform for observatories in the future.