Twice a year, a group of experts release a ranked list of the world’s most powerful computers called TOP500. It is likely that the new list in June will have a new member of the Top 10 of the Top 500: a computer dubbed Pangea. Its output is is 2.3 petaflops. A petaflop is a quadrillion “floating-point operations per second.” Today’s desktop computers deal in gigaflops, or billions.
The system is the fastest commercially-owned computer in the world. The other faster computers on TOP500’s list are owned by governments or academic institutions and therefore used for research.
Pangea is owned by Total SA, the fifth-largest oil and gas company in the world. So the supercomputer will not be changing the future of health care IT like former Jeopardy champion Watson or revolutionizing climate projections and weather research like supercomputers at NCAR and Oak Ridge National Laboratory. It will be searching for oil and gas, according to Reuters.
Pangea helped analyze seismic data from Total’s Kaombo project in Angola in just nine days, instead of the four and a half months it would have taken with its previous computer, Philippe Malzac, IT director at Total’s Exploration division, told Reuters:
Total trumps British rival BP with the 2.3-petaflop supercomputer. BP said last December it was building a 2 petaflop supercomputing facility in Houston, Texas.
“Our competitors are also working on these kind of algorithms, but we think this is giving us a head start,” Malzac said.
The price of the system is undisclosed, but it will cost nearly $20 million per year just to run Pangea. The technological achievement may be impressive, but the reality is that oil and gas reserves are finite and getting more expensive to extract, while renewable fuels like wind and solar are getting cheaper to utilize.
Raymond T. Pierrehumbert, a lead author on the third IPCC Assessment Report, explained last month in Slate that it is getting harder and more expensive to squeeze oil out of the ground.
Oil production technology is giving us ever more expensive oil with ever diminishing returns for the ever increasing effort that needs to be invested. According to the statistics presented by J. David Hughes at the AGU session, we are now drilling 25,000 wells per year just to bring production back to the levels of the year 2000, when we were drilling only 5,000 wells per year. Worse, the days are long gone when you could stick a pitchfork in the ground and get a gusher that would produce for years.
That is when an oil company knows where to drill without the help of a historically fast supercomputer. Global oil and gas exploration and production costs are expected to rise again to $644 billion in 2013, according to an annual survey by Barclays. These fuels are getting increasingly expensive and difficult to produce, requiring massive computational power to find a way to squeeze more dinosaur juice out of the Earth’s crust. The climate clock is ticking, and it is worth asking if such investments in oil & gas extraction are worth it. Pierrehumbert again puts the scenario plainly:
Whales were driven to the brink of extinction before petroleum replaced whale oil, and we may well fry our planet—and bankrupt ourselves while doing so — before we’re finally forced to kick the fossil fuel habit. It will be hard to muster the resources to develop replacements for fossil fuel energy if we wait until both the economy and climate are in ruins. We are in for a hard landing if we don’t use our current prosperity to pave the way for a secure energy and climate future.
That includes using recent powerful technological advances to get ourselves off fossil fuels and onto renewables.