AWS for Industries
Accelerate the Seismic Data Workflow
Seismic data, and the accompanying subsurface image and interpretation, are some of the primary assets for oil and gas operators with active exploration and development teams. Quick, seamless access to seismic and interpretation data from various software applications is a key component of all integrated seismic workflows for every exploration team at all oil and gas companies. When a geoscience interpreter spends more than half of their time to find, procure, and convert seismic data to the applications, this affects workflow cycle time and interpretation quality that results in suboptimal reserves estimation.
In this post, we explain the various components of the AWS Seismic Workflow Optimization solution that improves exploration teams’ productivity by speeding up data discovery and selection, minimizing manual data movement, format conversions, and seismic volume duplication between applications and systems. Let’s dive deep into how AWS can benefit oil and gas customers.
AWS and AWS Partners have created a solution that addresses the three primary challenges surrounding seismic interpretation that affect the whole industry. They are:
- Non-productive time due to data management tasks
- Inefficient software usage due to data processes that lock applications
- Complex multi-vendor workflows limit the innovation adoption of new solutions and technologies
The first challenge is the excessive non-productive time caused by the copying, downloading, converting, and conditioning of complex, large seismic data. This challenge is exacerbated by the fact that tens to hundreds of copies can be generated for each seismic dataset that enters an interpretation workflow, due to duplication, format conversions, and parsing of the original dataset. These highly inefficient seismic interpretation workflows not only negatively impact storage space and cost, but more importantly are a major contributor to non-productive time across the exploration and development organization overall. An estimated 60% of the overall workflow time is spent copying, opening, converting, and exporting seismic data.
The second challenge is the unnecessary lock of expensive software and related software license costs. Geoscientists utilize an average of $200,000 yearly software licenses from multiple vendors, during the interpretation workflow. These software licenses are locked, and unable to perform other functions, during seismic file operations; optimizing or completely removing these costly file operations will translate to an associated 60% software licensing optimization.
The third challenge is around the complexities of the interpretation workflow as it involves several applications from various vendors, limiting customer’s ability to optimize their workflow, and reducing non-productive time with the adoption of modern cloud computing. By moving to AWS and implementing modern cloud workflows, energy customers are reducing IT infrastructure costs and deriving more value from their seismic data.
The AWS Seismic Workflow Optimization solution enables energy customers to solve these issues.
The following reference architecture shows the overall Seismic Workflow Optimization solution.
Figure 1: Seismic Workflow Optimization Reference Architecture
The solution starts with a full resolution OSDU compliant seismic data lake that is hosted on Amazon Simple Storage Service (Amazon S3). Seismic data is traditionally stored in separate storage systems to cover data archiving, seismic processing, and seismic interpretation. Often, all of these storage systems have to be redundant to protect from hardware failures and data loss. Amazon S3 is intrinsically redundant across multiple Availability Zones by design and provides S3 Intelligent-Tiering to deliver optimized performance and costs based on dynamic business needs. The combination of S3 Intelligent-Tiering, redundancy, and adaptive performance addresses the needs of all the use cases across the seismic data workflow.
In the Seismic Workflow Optimization solution, the data is stored on Amazon S3 in the OSDU’s OpenVDS format, a lossless, cloud native format that supports seismic data from acquisition to reservoir. OpenVDS gives customers the ability to adopt a modern standardized format that provides high-performance data access and in addition can be streamed to subsurface applications across the workflow.
Seismic data, typically in SEG-Y format, can be moved to the cloud using AWS DataSync, AWS Snowball Data Migration, or individual customer uploads through AWS Direct Connect, and then ingested into OSDU. The data can be pre-converted to OpenVDS before transfer to the cloud or it can be converted once it arrives in the cloud leveraging the OSDU OpenVDS conversion or the Bluware Teleport tools.
The applications are installed and run on Amazon Elastic Compute Cloud (Amazon EC2) instances together with NICE DCV, a highly performant, no-cost, remote visualization protocol, enabling interpreters to access the software remotely. These compute instances are scalable and optimized for seismic workflows. The applications share local files using Amazon FSx, and Amazon RDS for their storage needs.
With the seismic data stored in OpenVDS format on the data platform, Bluware FAST, is used to automatically transcode from OpenVDS to the required seismic data format consumed by the various geoscience applications (at this moment support exists for SEG-Y, ZGY, PS, SEP, and a proprietary format from an oil major). Bluware FAST is installed on the Amazon EC2 instance where the applications run and performs two functions; first, it maps the OpenVDS seismic data from Amazon S3 into local virtual files in the format selected by the user; second, it streams the data on demand from Amazon S3 to the local disk. The mechanism is transparent to the interpretation applications as they will continue to consume SEG-Y or their own proprietary format.
Bluware provides a way for any company to add support for its proprietary format to FAST using an API to protect a company IP related to its data format.
The interactive data transcoding and streaming completely removes the need to copy and convert large seismic data files to the EC2 instance where the interpretation software runs.
On the other hand, modern cloud native solutions, and strong OSDU partner, INT IVAAP, already supports OpenVDS natively and can visualize the data in OpenVDS format directly from the data platform, and it can be used to browse and search OSDU data. All OSDU partner solutions, including Eliis Paleoscan, Geoteric, and Ikon Science RokDoc, demonstrate the value of leveraging a common format across the seismic workflow with its seamless integration with OpenVDS using seismic data streaming. This value continues to increase as other vendors are looking at OpenVDS as a format it will support in the future.
To close the loop and keep the data platform consistent, when any application generates new seismic file on the local Amazon FSx file system, it is detected automatically and an AWS Lambda job converts the newly generated seismic data from the application’s proprietary format to OpenVDS. Once the new OpenVDS files are generated, they are re-ingested into the OSDU data lake for use by other applications in the workflow.
OSDU’s OpenVDS has additional benefits for computational algorithms and applications and for AI/ML applications, as the streaming data model has even more value for those use cases. Therefore, this is just the first step in a high-value journey that will allow oil and gas companies to adopt cloud computing to transform workflows, increase efficiency, and reduce time to decisions.
In this blog, we described key benefits of the AWS Seismic Workflow Optimization solution to oil and gas customers operating into the exploration and development space. The solution uses a centralized seismic data platform to simplify the ingestion of a standardized data format. It conditions the seismic data on the fly and streams the data into multiple software and tools used throughout the seismic workflow on-demand. It gives customers the opportunity to move seamlessly between steps in the workflow and reduces or eliminates import and export steps. It unlocks the capability for customers to focus on key business objectives rather than data management and avoid costly non-productive personnel and software down time.
We also presented the overall reference architecture for the AWS Seismic Workflow Optimization solution in this blog. Specifically, we discussed how this solution can provide self-service seismic data management, conditioning, and streaming into other subsurface interpretation tools.
To learn more about how you can transform the core and build the future of your energy business, see AWS Energy.