by Osama Qazi
Modern seismic interpretation and visualization applications require high-end cutting-edge resources to enable users to take full advantage of the applications’ vast potential. Applications in this space now are engineered to operate on the Microsoft® Windows® platform as well as UNIX-based operating systems, and thus take advantage of the latest operating system features, high-end processing power, and huge amount of memory that are available today on powerful workstations. Such applications enable users to analyze and interpret more data and to do so in greater detail. The architecture can provide great benefits to users while simultaneously introducing new and significant data management challenges.
Data for seismic interpretation and visualization is a significant consumer of storage, and it can account for the largest amount of data storage needed outside of seismic processing. With greater file sizes and I/O generated for read or write to hundreds of files across the network, not to mention its many users, the network as well as the legacy file servers become saturated to the point at which data access becomes a big problem. In addition, concurrent access by the hundreds of tools used in the various phases of the workflow adds to the diverse mix of I/O requested from the storage infrastructure. Moreover, it's not just UNIX-based network access (NFS) that is required, but also Windows file sharing protocol (SMB) for the Windows-based applications.
Figure 1) Simplified exploration workflow process and data flow.
These and other factors become bottlenecks in the workflow process and inhibit users from efficiently completing their 3D structural models. The architectural changes in the network and storage that are required are time consuming, require research, and, most importantly, need huge investments in hardware and data center space. Therefore, in many cases, users have adopted the practice of saving their project data as well as seismic information on local disk. This alleviates some of the bottlenecks associated with networked storage but again introduces challenges related to data reliability, data integrity, and, most importantly, data protection and security. This is a big challenge, but not the only one. Geoscience users today have to recreate most of their work in order to go back to a previous or a different version of their interpretation. Naturally, this results in many saved copies of different versions of their interpretations on local disk. This can become a data management nightmare, especially when there are many users.
NetApp is the number-one data storage provider for upstream oil and gas. And the NetApp® clustered Data ONTAP® operating system is the operating system that enables the incredible features that help solve the most difficult data management problems faced by upstream oil and gas data centers. Clustered Data ONTAP provides:
- Performance—Data ONTAP’s efficient architecture, optimized code, and high-end hardware deliver the performance that's required. The operating system supports the fastest networks and the fastest storage medium with intelligent algorithms to cache frequently accessed data blocks at the faster tiers, eliminating trips to the spindles on every read.
- Nondisruptive Operations—The clustered approach provides the ability to completely eliminate planned downtime. No downtime is needed for software or firmware upgrades, hardware upgrades, or expansion. You can run your business 24/7/365!
- On-Demand Flexibility—Only clustered Data ONTAP provides the capability to make changes to the cluster size by adding or removing nodes or storage shelves, increase or decrease volume capacities, and, uniquely, move workloads around to increase performance or reduce cost. Thus, it’s possible to keep large seismic read-only data on inexpensive SATA while maintaining performance for the visualization data on faster SAS or SSD.
- Operational Efficiency—Efficiency is built in even on the primary storage. This allows you to scale your infrastructure according to business growth while keeping existing IT staff. When data lives on the right $ tier, there is no need for complex migration scans and policies. Block-level deduplication of data keeps capacity from being wasted. Space gained after the deduplication of data is given back to the pool to be used for other needs.
- Snap Creator Framework for Application-Aware Backups and Restores—Build application-centric backup solutions using comprehensive and integrated clustered Data ONTAP technologies, including Snapshot™ copies and SnapMirror® and SnapVault® technologies.
- Unified, Multiprotocol, and Ubiquitous Access—Cater to both UNIX® (NFS) and Windows (SMB) file shares without needing to store and manage the data twice! Ubiquitous access allows data to be presented to all users from the same share, regardless of whether the data is situated on a different controller or different disk subsystem. The cluster-wide name space provides such access not only to UNIX and Windows shares but also to SAN clients, for example, databases or virtualized environments.
Figure 2) Storage architecture for exploration workflows based on NetApp clustered Data ONTAP.
Managing data and storage with clustered Data ONTAP from NetApp helps upstream oil and gas businesses to decrease the cost associated with overprovisioned, complex, and isolated storage infrastructures that do not deliver the quality and performance required to run efficient workflows. Clustered Data ONTAP provides a streamlined, efficient, always-on, flexible, and scalable architecture that empowers scientists to focus on finding hydrocarbons and not waste time on complex data management.
© 2013 NetApp, Inc. All rights reserved. No portions of this document may be reproduced without prior written consent of NetApp, Inc. Specifications are subject to change without notice. NetApp, the NetApp logo, Go further, faster, Data ONTAP, Snap Creator, SnapMirror, Snapshot, and SnapVault are trademarks or registered trademarks of NetApp, Inc. in the United States and/or other countries. Microsoft and Windows are registered trademarks of Microsoft Corporation. UNIX is a registered trademark of The Open Group. All other brands or products are trademarks or registered trademarks of their respective holders and should be treated as such.