Abstract
Big data has attracted an increasingly number of attentions with the advent of the cloud era, and in the field of seismic exploration, the amount of data created by seismic exploration has also experienced an incredible growth in order to satisfy the social needs. In this case, it is necessary to build a highly-effective system of data storage and process. In our paper, we aim at the properties of the seismic data and the requirement to the performance of IO, and establish a distributed file system with the goal of processing seismic data based on the Fast Distributed File System (Fast DFS), then test our system through a series of operations such as file write and read, and the results show that our file system is very proper and effective when processing seismic data.
Highlights
When coming into the year of 2013, the term big data[1] has been mentioned much more times than ever before following after terms the Internet of Things, cloud computing[2]
The remainder of this paper is organized as follows: Section 2 describes the format of seismic data and our changes according to this kind of format; Section 3 details the architecture of Fast DFS and its working theory, and we add our work into this file system to create a new file system in order to satisfy the needs of seismic data; Section 4 tests our new file system with a series of operations and compares the performance with other file systems; Section 5 concludes our work
Fast DFS[11] is a open-source distributed file system which is similar to Google File System (GFS), it is accomplished by pure C language, supporting many UNIX operation systems such as Linux, FreeBSD, AIX and so on, it can only be accessed by proprietary API
Summary
1,2,3Department of Automation,CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System, University of Science and Technology of China, Hefei, China. Big data has attracted an increasingly number of attentions with the advent of the cloud era, and in the field of seismic exploration, the amount of data created by seismic exploration has experienced an incredible growth in order to satisfy the social needs. We aim at the properties of the seismic data and the requirement to the performance of IO, and establish a distributed file system with the goal of processing seismic data based on the Fast Distributed File System (Fast DFS), test our system through a series of operations such as file write and read, and the results show that our file system is very proper and effective when processing seismic data
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have