Do we really need exascale computers? Geophysicists say yes, we do

Are there any industrial problems requiring supercomputers 1000 times more powerful than those that currently exist? The answer from industrial partners of the HPC4E project, and specifically from the Workpackage of Geophysics Exploration for Hydrocarbons, is a definite “yes”.

Nowadays, HPC needs in geophysical exploration for hydrocarbons are almost all concentrated in subsurface imaging, focusing on the most demanding application in terms of computational resources. Although the imaging process derives directly from seismology, it also includes knowledge and techniques from many other scientific disciplines. In very approximate terms, subsurface imaging consists of a complete ultrasound of a huge area of exploration interest.

Despite of the existence of commercial suites for subsurface imaging in the market, oil & gas leaders have invested years in developing proprietary software and workflows. These tools and the technical results they have facilitated, literally considered the “crown jewels” by such companies, are the basis for multi-million dollar decisions, with a clear impact on the company’s income.

Mathematically speaking, imaging corresponds to the resolution of an inverse problem; processing it requires months of work by a multidisciplinary team of experts in mathematics, geophysics, geology and IT. During this process, the most powerful non-academic supercomputers are required from the industry to run algorithms. Which part of the process requires most computational time? More than 95% of the time, the rigorous simulation of the physics of mechanical wave propagation.

As an initial conclusion, the competitive advantage for an oil & gas company is that having 1000 times more powerful computers means a reduction of the computational time to process the raw data of a seismic acquisition. For any company, time is money.

While the process of software migration and introducing new physics in our suites to new supercomputing facilities is very challenging, we are aware that the new computing capabilities will provide results which enable faster decision-making by senior management, resulting in new achievements for the company. In this new era, all stakeholders agree that, in the short term, the new capabilities will be used for two long-sought objectives within the common goal of obtaining an improved knowledge of the subsoil. The first relates to the effective introduction of uncertainty in geophysical data modelling, and the second to the joint inversion/interpretation of different geophysical information (such as seismic and electromagnetic data).

From the research point of view, we are convinced that the exascale era will open up new ways for HPC facilities to support geophysical investigation, ways that we still are not entirely able to visualize. For a company, the best strategy for early adoption of new technology, risk reduction and quick pay-off is to be part of a collaborative environment with non-competitive academics who are at the leading edge of science.  


Santiago Fernández - REPSOL


This article also appeared in the LinkedIn GroupJoin us!

Other LinkedIn articles:

Link Wind energy in HPC4E
Link Investigation of flame structure of biomass-derived gaseous fuels
Link Finding the "not so easy" oil
Link How supercomputing can help improving the energy sector
Link Effects of fuel composition on biogas combustion in premixed laminar flames
Link New generation subsurface imaging gets a boost from HPC
Link Novel Hybridizable Discontinous Galerkin method paves the way to tackle realistic 3D problems in seismic imaging
Link Improving short-range wind intensity prediction based on multimodel meteorological ensemble forecasts and Genetic Programming
Link Biogas Utilisation for Sustainable Power Generation
Link Preparing the oil and gas industry for the Exaflop era
Link Dynamical and statistical high resolution downscaling approaches for the surface wind
Link ​​Innovative multiscale numerical algorithms for highly heterogeneous media extended to seismic problems