Wall-to-wall machine learning for CGG GeoSoftware

PowerLog and Hampson Russel now expose open source ML functionality. CGG ‘lifts-and-shifts’ to the Azure cloud.

CGG reports extensive use of machine learning across its software portfolio. ML is used for a variety of ‘mundane’ tasks such as modeling missing log curves and identifying and flagging poor-quality data. The latest (9.7.2) release of its PowerLog petrophysical analysis software offers native machine learning and deep learning Python utilities, leveraging open-source technology in custom workflows. Likewise, the 10.4 release of HampsonRussell adds ‘substantial’ new ML functionality in its Emerge attribute prediction module. This uses a ‘deep feed forward neural network’ to the challenging task of density estimation in seismic inversion.

CGG is migrating all of its software products to the Cloud, first to Microsoft Azure and then to other cloud providers. An initial ‘lift-and-shift’ port is complete, now the company is working to scale-out CPU-intensive computations to take full advantage of the HPC capability of the cloud. CGG’s new CEO Sophie Zurquiyah said, ‘We are taking the lead in developing new workflows and capabilities that leverage the full potential of ML and the cloud.’ More from CGG GeoSoftware.

Click here to comment on this article

Click here to view this article in context on a desktop

© Oil IT Journal - all rights reserved.