Design And Implementation Of Ship Inspection Platform Of Maritime Safety Administration Based On Cloud Environment

Cloud environment refers to the Internet or big data environment that can provide users or various application systems with computing capabilities,storage capabilities,or virtual machine services on demand from a dynamic virtualized resource pool.The traditional maritime safety bureau ship inspection platform is usually deployed on a local server,and the service performance provided by the platform is also restricted by the hardware configuration of the server.The traditional maritime ship inspection platform only provides some simple functions of ship data information maintenance,certificate entry and inquiry,the law enforcement records of ships are usually organized and stored in paper form.With the increase in the number of ships,law enforcement officers have to deal with a large amount of ship information every day.The number of documents produced by law enforcement is also extremely large,it’s hard to cope with traditional storage methods.The efficiency of the existing Maritime Safety Administration’s ship inspection platform has obviously not kept up with the development needs of the modern shipping market.Thesis combines the advantages of the cloud environment to design a set of maritime safety bureau ship inspection platform based on the cloud environment.The platform is deployed on the cloud server to avoid the low performance of the platform caused by server hardware configuration.The data generated by the ship inspection platform of the Maritime Safety Administration is mostly small files,the traditional HDFS usually exhibits low performance when storing massive small files.In order to solve this problem,thesis mainly does the following work。(1)Thesis conducts in-depth research on the existing small file storage optimization model.Combining the characteristics of the close relevance of ship inspection data of the Maritime Safety Administration,a small file merging method based on Map File was proposed.Use <key,value> pair as a record and store in data,realize the merging of small files to large files by serializing the <key,value> pair,at the same time,establish a mapping from small files to large files,the key value of each record and the offset of the record in the file are stored in the index.When the file is accessed,the location of the file where the record is located can be quickly found according to the index information.(2)On the basis of the traditional architecture,a department node is added as an intermediate processing module.Before the file is uploaded to HDFS,it will go through the department node to preprocess the uploaded small file,store the processed file metadata information in the Name Node.After merging small files,design a reasonable file index structure,and store the mapping relationship between small files and large files in HBase.Set up a cache prefetch module to cache file metadata corresponding to hot data,and optimize file storage and reading processes.After the preprocessing of the files by the intermediate processing module,the performance of HDFS processing massive small files is significantly improved.(3)Detailed design and implementation of the platform according to the actual needs of the Maritime Safety Administration,and thesis tested the designed ship inspection platform,the results show that the functions of the Maritime Safety Administration ship inspection platform based on the cloud environment meet the actual needs of the Maritime Safety Administration.Therefore,the ship inspection platform of the Maritime Safety Administration designed in thesis has strong practical application value.