Petrobras Uses The Largest Supercomputers In Latin America To Optimize Oil And Gas Exploration And Production Activities

With the Ranking Update, Atlas and Fênix, Also from Petrobras, Are the Two Largest Supercomputers in Latin America

In order to have a fully digitized company, to accelerate production time, Petrobras uses a new supercomputer to optimize oil and gas exploration and production activities. Petrobras and Açú Petróleo Sign Contract That Could Double Flow at the Port of Açú

Read Also

• Nuclear Plant Angra 2 Will Be Shut Down for 22 Days
• Construction Work Demands Job Opportunities for Pipefitters in MG and SP
• Welders, Technicians, Engineers, Supervisors, Inspectors, and More Job Opportunities in Mining Projects on This Day 23

About 1.5 million smartphones or 40,000 high-end laptops. The processing power of the Atlas supercomputer is equivalent to these devices combined.

Recomendado para você

Oil and Gas

Major Oil, Gas, and Energy Event Will Take Place in Brazil: Macaé Energy Will Gather Petrobras, Equinor, Prio, Among Other Suppliers and Energy Executives for Business, Networking, and Employment Opportunities

The Technical Agenda and the Fair in the City of Macaé Should Boost Discussions on Investments, Energy Transition, and Development of the Oil, Gas, and Energy Chain in the National...

Paulo Nogueira

1

In operation since April, at the Data Processing and Information Treatment Center (CPTI), in Rio de Janeiro (RJ), Atlas ranks 57th in processing capacity on the Top500.org list, published on Monday (06/22). The supercomputer is the highest-ranked in all of Latin America on the global list of high-performance computers.

Prior to this, in the list published in June last year, the Fênix supercomputer, also from Petrobras, held the top position. It underwent an update this year and now ranks as the second most powerful in Latin America, at position 83 on the list.

According to Petrobras, Atlas and Fênix are part of a team of several supercomputers that perform different functions within the company. The two high-performance computers – or HPC, for short – are responsible for processing geophysical data generated during oil and gas exploration and production development activities. Together, they have processing power equivalent to that of 2.5 million smartphones or 67,000 new laptops.

This processing power is necessary to create representative images of the geology below the seabed, where the salt layers and, of course, the oil reservoirs are located.

The seismic images, crucial for oil and gas discoveries, cover hundreds of square kilometers and reach depths of thousands of meters. Therefore, the algorithms that process them involve complex mathematical equations, dealing with a massive volume of data, generating images that geologists and geophysicists can interpret.

The data volume for a single seismic project can reach dozens of terabytes, exceeding the capacity of a current desktop computer’s hard drives.

Investment in HPC reduces costs associated with this type of processing. Additionally, with dedicated supercomputers, it is possible to utilize special algorithms developed by Petrobras, providing a competitive edge capable of enhancing the efficiency of exploratory activities.

Petrobras Strengthens Investment in High-Performance Computers

Since 2018, Petrobras has increased its investment in high-performance computers. This year, the company also began using Atlas and Guaricema – dedicated to simulating data generated in oil and gas reservoirs.

As a result of this investment, there has been a significant reduction in the time taken to process geophysical data in the geological areas of interest to the company and the application of complex, state-of-the-art algorithms.

This faster processing with more efficient algorithms will reduce risks and accelerate decision-making, boosting the economic return of E&P projects.

Therefore, the arrival of Atlas and the improvement of Fênix will contribute to better technical decisions, as the images generated below the seabed will have better definition and will be available in less time.