Business Bottom Line—Optimizing performance: Object storage for seismic applications

Abstract

Political instability at many points of the world … rising gasoline prices … limited natural gas and oil reserves. These are just a few of the challenges facing the oil and gas industry as it attempts to keep pace with the energy demands of an increasingly developed world. In this quest to supply the globe, it's no surprise that a growing number of energy companies rely upon technological advances to increase their efficiency and productivity. Nowhere is this more evident than seismic data acquisition and processing capabilities, where 4D seismic and other advanced imaging techniques are giving geophysicists more opportunities to accurately analyze the earth's subsurface to make predictions on where and when to drill. Complementing these recent developments has been the advent of Linux compute clusters, which give affordable and extremely powerful computing capabilities to handle the demands of these advanced imaging techniques. These clusters have forever changed the high-performance computing landscape but the storage capabilities have not kept up with the cluster's demands for massive data bandwidth. But today, an object-based storage architecture offers a solution in which the storage system's scalability and price-performance can match that of the Linux clusters. Current network storage systems have been simply incapable of providing the data throughput needed to keep Linux clusters operating efficiently. Each of the two major types is distinguished by its command sets. …