Database And Transmission Protocol For Seismic Data Acquisition

A cloud-based data storage system and a low-latency transmission protocol for seismic data acquisition are studied and implemented in this thesis for handling the performance issues encountered in the "Large-scale images,surface and seismic visualization and performance optimization" projects.Single computer could not offer satisfied performance in processing massive seismic datum,so we developed the new data storage system and network protocol for the establishment of cloud computing platform,which can deal with large-scale data.The seismic data storage system is designed for storing surface and seismic data.With cloud database as underlying data warehouse,it can store massive seismic data.It uses cache to solve the additional overhead introduced by the remote database.The Least Recently Used(LRU)cache is implemented by a linked list and a hash table,which improve the storage performance further more.The test results show that the access latency performance of the data storage system is significantly improved compared with the previous file system,which ensures that the rendering program can retrieve the data from the cloud database in real time.The network transmission protocol is designed for transmission of the 3D surface and seismic images rendered in the seismic data visualization system.The protocol adopts optimized data confirmation and retransmission logic to achieve reliable transmission with low delay.Test data shows that the protocol provides a lower transmission delay than the TCP protocol in similar network circumstance.This protocol plays an important role in the data visualization system based on cloud computing.