On this side of the ocean, Cai Siqiang was busy with the acquisition, but Duke did not consider this matter at all. He now thought about how to get the nuclear reaction simulation program in the nuclear laboratory, except for dealing with the course every day.

This matter is too difficult. So far, Duke has only been able to use nuclear reaction simulation programs through one terminal desktop application. The most valuable asset of MIT Nuclear Laboratory, the establishment of its model almost focuses on the data calculation results obtained by the nuclear laboratory and other nuclear reaction research institutions in the previous nuclear tests, most of which also use subcritical tests in the laboratory to verify the simulation calculation results.

It can be said that this application focuses on the research of countless scientists in American universities and official nuclear physics research. Although Duke does not know if there are similar systems in China, even if there are, I believe there is a big gap compared with this system. After all, the research foundations of both sides are different. Whether it is computer modeling or experimental verification, the gap between the two sides is not a small part.

Moreover, even if there is such a simulation system in China, Duke will never have any chance to contact. Systems like this are basically top secrets unless Duke decides to be an invisible person.

Therefore, this nuclear reaction simulation program from MIT is Duke's only hope.

But this hope seems a bit slim.

Duke cannot find a way to break through the nuclear reaction simulation application system. Kerry has seen the effect of streamlining the new operating system. The system streamlined on the basis of crashlinux has been successfully run in the simulated 2-channel 20-core workstation platform.

Compared with the Linux system originally brought by this workstation, Kerry's streamlined and optimized new system has doubled its performance. What is particularly rare is that it retains 100% compatibility with the original Linux system on this basis, which means that the original application can be run directly on the new system without any changes.

This new system is reduced to half of the original code, and the compiled binary code has been reduced by almost forty%.

In the following further simulation tests, the 4-channel and 8-channel simulation systems were also verified, proving that Kerry, a streamlined system, has at least fully implemented the concept of a new layered architecture in terms of function.

Next, Kerry further expanded this system to other CPU platforms to see if the effect of this new architecture is as good as other CPU platforms.

Seeing Kerry's outstanding performance in operating system improvement, Duke suddenly felt aroused and instructed Kerry to optimize and improve the operating system of the Titan Ii prototype to see if he could achieve some results. Since Kerry is already very familiar with this platform, the results were soon released. After the optimization and improvement of the Titan Ii prototype, the new crashilinux system was increased by nearly 50%.

This optimization effect scared Duke. This is a supercomputer. It is like upgrading the system again without spending money. Duke called out the Kerry optimization and improvement and found that the biggest improvement comes from the improvement of the parallel scheduling algorithm. Just this place, modifying thousands of lines of code will directly improve the system performance by about 15%.

"Duke, can you stay with your modified program? We would like to contact Cray to see if you can use the improved operating system to improve the Titan II." Andrew solicited. This improvement is a result of Duke after all, and a very important result. Andrew certainly needs to obtain Duke's permission to source code, although Andrew is also Duke's mentor.

Although Duke is a bit painful for taking this improvement, Cray will be able to use this result in the future on its new models. According to the expensive cost of supercomputers, this new algorithm that has increased by up to 15% is worth billions. But now it has to be taken out because Duke hopes to improve this scheduling algorithm to lift out the nuclear reaction simulation program.
Chapter completed!