Gazprom Neft Develops Supercomputer To Simulate Siberian and Arctic Fields
Gazprom Neft has built a computing cluster in St. Petersburg, Russia, designed to generate digital twins of oil fields. With an intranode data-transmission rate exceeding that of many Russian supercomputers, the new distributed-computing system is capable of processing more than 100 gigabits per second, speeding up the digital-modeling process five-fold.
Digital field models help specialists make informed investment decisions, plan asset infrastructure, and estimate hydrocarbon production volumes. Computer processing of 3D models requires analysis of many physical, engineering, and economic variables. To that end, Gazprom Neft’s new project uses cluster parallel computing, in which all tasks are divided into several concurrent processes, increasing speed and efficiency.
The computing cluster has been used successfully by specialists at Gazprom Neft’s Science and Technology Centre in preparing calculations for developing the Vyngayakhinskoye and Shinginskoye fields. The supercomputer has demonstrated a performance of at least 16 teraflops (16 trillion floating-point operations per second) in pilot testing; this capacity is equal to 160 conventional computers.
Intel’s OmniPath product forms the main data-exchange network, with video cards being used in all calculations, increasing processing speeds. Using these technological innovations has made it possible to gain an advantage in addressing the tasks involved in developing hydrodynamic field models.
As well as increasing speeds in developing 3D models, the Gazprom Neft computer cluster can create these models in more detail and more accurately than others based on systems currently available on the market.
The new cluster’s computational capacity is expected to be used with fields in the Orenburg region and Sakhalin soon. Thereafter, the project is expected to be used to improve efficiency in developing Gapzrom Neft’s assets in the Arctic.
Don't miss out on the latest technology delivered to your email monthly. Sign up for the Data Science and Digital Engineering newsletter. If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.
21 October 2019
16 October 2019
16 October 2019