A group of Swiss researchers claims that their high-performance computer has added 12.8 trillion new digits to the number Pi, resulting in a total calculation of 62.8 trillion figures.

The high-performance computer, housed at the University of Applied Sciences of the Grisons’ center for data analytics, visualization, and simulation (DAViS), completed the Pi calculation with a precision of exactly 62,831,853,071,796 digits, smashing Timothy Mullican’s previous record of 50 trillion digits set last year.

Prior to Mullican, the trophy was held by Google, whose team discovered over 31.4 trillion digits for Pi in 2018.

The Swiss team achieved the result in just over 108 days, which is three and a half times faster than Mullican’s previous record of 303 days, and it is now awaiting verification before being entered into the Guinness World Records. The entire number will not be made public until then, but the researchers teased that the last ten known digits of Pi are now: 7817924264.

For most people, the number Pi conjures up images of math class, where it was defined as the ratio of a circle’s circumference to its diameter and frequently shortened to its first few digits: 3.1415.

Mathematicians have been attempting to calculate the digits of Pi with as much accuracy as possible for centuries – in fact, as far back as the ancient Babylonians. However, because Pi is an irrational number, which means it can never be represented with absolute precision, the point isn’t to find practical applications; rather, the calculation has become an unofficial benchmark for high-performance computing, as well as an opportunity for scientists to compete against one another.

The DAViS researchers used a well-known algorithm known as the Chudnovsky formula, which was developed in 1988 and is widely regarded as the most efficient method of calculating the number Pi. The Chudnovsky algorithm was also used by Google’s team and Mullican.

The algorithm was executed using y-cruncher, a popular computer software program created in 2009 by American developer Alexander Lee specifically to compute Pi.

According to the Swiss team, one of the most difficult challenges was the amount of memory required to perform such a large calculation. DAViS’s high-performance computer was outfitted with two AMD Epyc 7542 processors and 1TB of RAM, which was insufficient to hold all of the digits they hoped to generate. As a result, the y-cruncher program was used to swap out the digits to an additional 38 hard disk drives (HDD) with a total storage space of 16TB, saving a large portion of the RAM on the HDDs.

The computer and disks could reach temperatures of up to 80°C during operation, so the system was housed in a server rack with constant air cooling to avoid overheating. This contributed more than half of the total 1,700 watts of power estimated by the scientists to be required for the full calculation, putting the system in 153rd place on the Green500 list.

The Chudnovsky formula, for example, is well-known for its complexity: when scientists implement the algorithm, they discover that the time and resources required to calculate the digits grow faster than the digits themselves, while it becomes more difficult to survive hardware outages as the computation grows.

The new achievement, according to the Swiss researchers, reflects the capabilities of high-performance computing systems and their potential for other research areas. “The calculation demonstrated that we are prepared for data and computing power-intensive use in research and development,” said Thomas Keller, project manager at the Grisons University of Applied Sciences. “The calculation also made us aware of infrastructure flaws, such as insufficient backup capacity.”

DAViS encourages the use of high-performance computing in machine learning, such as in the Translaturia project, which is developing a computer-aided tool to translate from the Romansh language, which is spoken primarily in the Swiss canton of Grisons and is on the verge of extinction.

The computing center is also investigating the use of DNA sequence analysis in allergy and asthma research, which necessitates the use of high-performance computing systems. The new record contributes to laying the groundwork for future practical applications.