Emptied of the laboratory knowledge reserves of the Department of Materials Science, Duke set his sights on the nuclear laboratory.

However, the accumulation of knowledge in the nuclear laboratory is not as simple as that of the laboratory of the Department of Materials Science. Because the essence of the nuclear laboratory is all in a behemoth.

When Duke first entered MIT's nuclear lab, another of Professor Andrew's Ph.D. students, Hewitt, was already waiting there.

"Dr. Duke, welcome to the nuclear laboratory, my name is Hewitt, Professor Andrew explained that I will take you to familiarize yourself with the laboratory environment today." Hewitt, from Illinois, USA, wears a gold-rimmed glasses, curly brown hair, and a casual denim outfit, looking more like an artist than a scientist.

"Thank you. I'm sorry for you today. Duke thanked graciously. Judging from Hewitt's expression, it is not enthusiastic, but it is by no means indifferent, treating Duke like an ordinary person, as if he has never heard of Duke just winning the Turing Award, he is indifferent, and he has more artist temperament than a scientist, this is Duke's first feeling about this Dr. Hewitt.

Compared to the laboratories of the Department of Materials Science, most of the nuclear laboratories do not see a variety of instruments everywhere, but they are very neat. But most of these are offices with computers, and Hewitt explains whose offices they are, including where Professor Andrew and his own office are, and even Duke has a separate office.

Duke didn't feel very different from the outside company office, and he couldn't see any difference at all.

Seeing that Duke seemed to be a little disappointed, Hewitt smiled slightly, "Actually, what I just looked at is just the office of the professors and doctors' daily work, so I will take you around today so that you can find them in the future." The place not far ahead is the highlight of our visit today. ”

Sure enough, after a while, Hewitt led Duke into a huge room, a large hall to be exact, but it was filled with dense cabinets.

"This is our computing center, where one of the most powerful supercomputer systems in the world today is located." Hewitt showed a rare little passion along the way, and at this moment he couldn't help but explain with some pride when he saw these cabinets.

"This is Cray's 2012 Titan II supercomputer system tailored for the MIT Computing Center, which is a new system born out of Cray's Jaguar supercomputer architecture, but equipped with a significantly smaller number of physical CPUs than the old version, using just over a third of the Jaguar.

However, due to the upgrade of the CPU from the original AMD Istanbul 6-core to the main 20-core Terramar processor launched by AMD in 2012, the number of real computing cores has increased a lot. From more than 1.2 million to more than 2 million, the performance has increased nearly tenfold compared to the jaguar, and the stable computing speed has reached 15 petaFlops (that is, 150 million petaflops per second).

The cost of the entire system is only $35 million, which is a good price.

In this system, there are now nuclear fusion reaction and nuclear fission reaction simulation systems used in our nuclear laboratory, as well as complex galaxy formation simulation systems in other physics disciplines, plasma fusion reactor simulation systems, and global climate simulation systems.

It's fair to say that this is the core of MIT's physics research, and it's the system that's the essence of our research. ”

Duke is the first time to see a real supercomputer up close, and he has seen China's Tianhe series of supercomputers from the news before, but that is what he saw on the TV screen, and he forgot about it after seeing it, and he didn't feel anything at all.

At this moment, I saw a real supercomputer in front of me, and the first feeling that came to my face was that it was large, large-scale, and even very large compared to the minicomputer computer system of Duke's original school, and the area of the hall of the computing center was preliminarily estimated, and Duke felt that it was at least 500 square meters, which was really worthy of a supercomputer, and this head alone made people awe-inspired.

"Dr. Hewitt, I don't see any screens or workbenches that can be manipulated, could it be that the interface is somewhere else?" Duke observed for a while, and finally saw that something was wrong.

"This is the computer room of the computing center, and as for the operation of the terminal, it is on the office computers of the professors and doctors." Hewitt replied.

"Oh, I see." Only then did Duke understand why every office had a set of computers, and it turned out that in addition to the office, there was also the function of directly operating the supercomputer system.

"Can my office computer be connected to the system now?" Duke then asked.

"I don't really know about this, but you are a student of Professor Andrew, and even if you don't have it for the time being, you can apply to the computing center. You see, the application process is here. Hewitt continued to lead Duke to an application guide posted on the wall.

The process is very simple, the computing center has a permission application system, Duke can apply on it, and then the supervisor approves, and then the laboratory director approves, and the computing center opens the authority through these two gates.

"How about it, do you have any thoughts after reading this supercomputer system?" Hewitt said to Duke with interest, it can be seen that this system is his love, and his originally indifferent temperament immediately became a little passionate when he arrived here.

"It's huge." Duke said honestly.

"Hehe, it's really big." Hewitt couldn't help but laugh when he heard Duke's simple comment on the words, "After you use it, you will find that its performance is far beyond your imagination, since the computing center updated the computer system and completed the installation of Titan II, the research results of our laboratory next year will be more than one-third more than before."

Among them, there are not only the improvement of R&D ideas, but also the credit of this supercomputer system, because the simulation speed is hundreds of times faster than the previous system.

The simulation test, which originally took a month to get the calculation results, can now be available in a few hours, which has greatly improved our R&D capabilities. If you were to choose the most popular person in Nukalab, I think most people would vote for Titan II. ”