About 1.5 million smartphones or 40,000 latest generation laptops. The processing capacity of the Atlas supercomputer is equivalent to these devices, combined. In operation since April, at the Center for Information Processing and Treatment (CPTI), in Rio de Janeiro (RJ), Atlas occupies the 57th position in processing capacity of the Top500.org list, released on Monday (06/22). The supercomputer is the best ranked in all of Latin America in the world ranking of high performance computers.

In the previous Top 500 list, of November 2019, the title of the largest supercomputer in Latin America was Santos Dumont, which returned to the top position after its processing capacity was multiplied by five with investments. Before that, in the list released in June last year, the supercomputer Fênix, also from Petrobras, occupied that position. It underwent an this year and today ranks as the second most powerful in Latin America, at number 83 on the list.

Byte oceans

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

So much processing power is needed to create the representative images of geology below the seabed, where the layers of salt are and, therefore, the oil reservoirs. The seismic images, fundamental to the findings on oil and gas, cover hundreds of square kilometers and reach thousands of meters of depth. That is why the algorithms that process them involve complex mathematical equations, with an immense volume of data, generating images that geologists and geophysicists can interpret. The volume of data related to a single seismic project may reach tens of terabytes, more than the capacity of the HDs of a current desktop computer.

The investment in HPC reduces the costs with this type of processing. In addition, with dedicated supercomputers, it is possible to use special algorithms developed by Petrobras, bringing a competitive advantage capable of increasing the efficiency of exploratory activities.

Data Tsunamis

Since 2018, Petrobras has reinforced its investment in high-performance computers. This process has been occurring in three waves. The first, last year, resulted in the acquisition of Fênix. This year, the company also started to count on Atlas and Guaricema - the latter dedicated to the simulation of data generated in the oil and gas reservoirs. With the completion of the third wave in 2021, the company will have 10 times more HPC processing capacity than it had back in 2018.

As a result of this investment, there was a significant reduction in the processing time of geophysical data in the geological areas of interest to the company and the application of complex, state-of-the-art algorithms. This faster processing and with more efficient algorithms will reduce risks and anticipate decisions, increasing the economic return of E&P projects. That is why the arrival of the Atlas and the improvement of the Phoenix will contribute to improve technical decisions, since the images generated below the seabed will have better definition and will be available in a shorter time.

The Director of Digital Transformation and Innovation at Petrobras, Nicolas Simone, points out that the processing capacity of these computers allows the development of strategic programs for Petrobras, such as “PROD1000” and “EXP100”, accelerating the increase in efficiency in the company.

“The intensive use of artificial intelligence and large computational capacities guarantee an expansion of the level of data processing that would be impossible to achieve without these technologies”, concludes the director.

Technical files

Atlas

Number of servers: 136

Mathematical Accelerator (GPU): Nvidia V100 / 32GB / SXM2

Accelerators per server: 8

Processor: Intel Xeon Gold 6240

Processors per server: 2

Cores per processor: 18

Velocity: 2.6GHz

Memory per server: 768GB

Internal network: InfiniBand EDR 100gbps

Operational system: CentOS 7.7

Theoretical performance: 8.0 PFLOPS Rpeak DP

Measured performance: 4.4 PFLOPS Rmax (or 4,376 TFLOPS Rmax)

Consumption (at maximum capacity): 546KW

Fênix

Number of servers: 360

Mathematical Accelerator (GPU): Nvidia V100 / 32GB

Accelerators per server: 2

Processor: Intel Xeon Gold 5122

Processors per server: 2

Cores per processor: 4

Velocity: 3.6GHz

Memory per server: 192GB

Internal network: InfiniBand EDR 100gbps

Operational system: CentOS 7.6

Theoretical performance: 5,372 TFLOPS Rpeak DP (or 5.4 PFLOPS Rpeak DP)

Measured performance: 3,161 TFLOPS Rmax (or 3.2 PFLOPS Rmax)

Consumption (at maximum capacity): 390KW