Computing has advanced significantly from the days of abacuses and slide rules. Long passed are the days where punched paper cards were used for large-scale automated data processing. As our achy Blackberry thumbs can bear witness to, today’s computers can quickly perform a variety of operations simultaneously with relative ease on the smallest of devices.

On a grander scale, the same computers that generate Hollywood’s special effects also play an active role in the petroleum industry. Or to see it another way, the same systems that can bring an imaginary Davy Jones’ Locker to the big screen are helping researchers explore below it in real life.

The computing power necessary to undertake both is monumental, especially in the arena of advanced seismic processing. However, scientists and engineers for BP will soon have a much faster system to be creative with once the company’s new high-performance computing (HPC) center opens later this month.

Pet a what?

Construction of the HPC center at BP’s Westlake Campus in Houston began in December 2012. When completed, the new facility will house the largest supercomputing complex for commercial research in the world. Designed to support data processing at a rate of up to 2.2 petaflops – more than 2,000 trillion calculations in a single second – the HPC center will be housed in a three-story 10,220-sq-m (110,000-sq-ft) facility. It will replace the company’s 30-yearold HPC center, which started with mainframes in 1993 and grew to become the world’s first commercial research center to achieve 1 petaflop of processing speed in 2012.

In addition to the advanced processing rate, the supercomputing system has a total memory of 1,000 terabytes and disk space of 23.5 petabytes – equivalent to 147,000 Apple iPods with 160 GB memory. Those iPods, if stacked vertically, would climb nearly five times higher than the Empire State Building.

“We will have the tools and capabilities to test concepts in a controlled environment that will allow the best ideas to emerge, which can then be scaled up to the next level of development,” said Andy Brayshaw, BP’s vice president of emerging and integrated technology, upstream technology. “We’re combining science with math to find the best solutions.”

Creative advantage

The new HPC center will serve as a worldwide hub for processing and managing huge amounts of geologic and seismic data from across BP’s portfolio. This will enable faster, more informed business decisions generated from clearer images of rock structures deep underground and a better understanding of hydrocarbon flow through “digital” rocks.

The center will play a key role in identifying potential exploration targets and informing business development decisions from the Gulf of Mexico (GoM) to Azerbaijan. According to the company, the added computing power helps teams work more efficiently and accurately than ever before, reducing both drilling risk and the costs and timetables of future exploration programs.

“When you can’t test your idea, you can’t be as creative,” said Keith Gray, manager of the HPC center. “More computing power means we can try more ‘what if’ cases. There are ideas we are still trying to develop. Having access to more computer power means we can be more creative.”

Brayshaw added that increased computer power brings time and money savings. “With some offshore wells costing hundreds of millions of dollars, it’s critical that we build models to understand the full range of subsurface scenarios,” he said. “By building 3-D models of reservoirs and simulating fluid dynamics in advance, we can save both time and money and increase the chance of a successful development.”

Competitive advantage

HPC has been vital to the advances made by BP in seismic imaging over the past two decades. According to the company, advances like the wide-azimuth towed-streamer seismic technology used in subsalt imaging have transformed the way data in the GoM and other major offshore basins are acquired and processed. These advances mean the demand for computing power increases rapidly. For example, the computing needs of BP today are 10,000 times greater than they were in 1999.

“An imaging job of today, if done 12 years ago, would take three to four months,” Gray said. “With the new system, it will be a few days. There are jobs we couldn’t do 12 years ago that we can do today.”

He added that the faster processing times afforded by the new supercomputing system will provide a real competitive advantage for the company. For example, the system will provide better and more efficient planning of effective seismic surveys.

Brayshaw noted that the new system also provides a recruiting advantage by ensuring that the best talent entering the work force will have the fastest systems to work on.

Built to last

The new facility was designed to withstand the peculiar weather elements that sometimes plague the Texas Gulf Coast. Building designers needed to account for a variety of climate-specific issues, from high winds and potential flooding from summer hurricanes to power loss due to winter ice storms.

For example, more than 500 truckloads of dirt were brought in and leveled to ensure that the building sits 1.2 m (4 ft) above ground level. Exterior walls are 8-in.-thick precast concrete panels, and the walls surrounding the computer rooms are 6-in.-thick cinder block.

In the new facility the HPC center resides on the third floor, with the second floor being designated for future growth. The building’s mechanical systems are on the first floor. Life expectancy for the new facility is 25 to 30 years.

In addition to enabling future growth, the HPC center features improved electrical and cooling systems that will reduce the power consumption by 30% over the current facility, according to the company. Electrical and cooling redundancies are in place to ensure uninterrupted power to the HPC center. Diesel generators will provide backup power when necessary, and enough diesel fuel is readily available to keep the generators running for a significant time until power is restored.