Posted September 12, 2018 05:53:22 A computer in Arizona is now able to perform the same task as a human at the same time, thanks to a new software program.
The software, called “Mighty”, is based on a design called “super-intelligent super-supercomputers”.
The design is based partly on the work of a team of researchers at the University of California, Berkeley, who are building a supercomputing cluster in the US called the Computational Neuroscience Centre.
Researchers say that this technology could have enormous benefits in fields ranging from neuroscience to medical diagnosis.
“Makes sense that a super computer would have a big impact on human health, because we have an abundance of supercomputers around the world,” said David W. Pfeifer, a computer scientist at the university and a co-author of the paper.
“There are currently supercomputational algorithms that are able to solve the problems of cancer and cancer imaging.”
The software is now available for download on the University’s website, which is a joint venture between the University and the National Science Foundation (NSF).
“We wanted to make it accessible to people in academia and the public, but we didn’t want to charge people for it,” said PfeIFers colleague David H. Linton, an assistant professor of physics at UC Berkeley and the paper’s lead author.
“We didn’t expect that people would want to be able to run it on their personal computers, and that’s a great example of why the research is important.”
The computer in question is based in the Computationally Neuro-Nanotechnology Lab at the National University of Singapore.
The researchers designed the supercomputer using two different computer architectures: a custom computer architecture called a “microprocessor”, and a standard architecture that is used in other supercomputed tasks, such as image analysis and data mining.
“The architecture we built allows us to have an efficient algorithm for a wide range of problems that is both fast and flexible, and allows us the flexibility to choose the right algorithm for the right problem,” said W. James Gartrell, an NSF postdoctoral fellow in computing at UC Davis, who is not involved in the research.
“For example, in a real-time image processing problem, you might want to get the same result for a variety of different images in a single image processing step.
That is a lot easier to do with a microprocessor, because you can just set the parameters.”
The researchers were also able to build the supercomputer in a way that makes it more compact.
“If you have the microprocessor working in a microcomputer, you get a lot of information, but if you have it in a computer, you only get one piece of information,” Gartrel said.
“But the architecture in this microprocessor allows you to work in a super compact manner.”
The supercomputer can do all sorts of calculations in parallel, so it can be used for many different tasks, said J.P. Trombino, a professor of computer science at the California Institute of Technology.
“It can solve some big problems, and it can do many more small ones, so we’ve got a lot more power in one computer than we’d have in a conventional microprocessor.”
Pfeife, who has been working on supercompositional algorithms for many years, was not surprised to see that a system like this could be so powerful.
“Our understanding of how supercomparatively efficient the problem-solving algorithms are is pretty good, but there are many other challenges that are harder to find,” he said.
The team is currently working on a system that can compute large numbers of operations per second, and the researchers are also looking at how to improve the efficiency of the system.
“You can use this supercomputer to do the same thing in parallel as you would in a traditional supercomputer, but you can also make the super computer do something that’s more efficient than a conventional supercomputer,” Pfeifers said.
P.J. Travino is a professor at the Computer Science and Artificial Intelligence Laboratory at the USC.
He is not part of the research team, but was a coauthor on the paper with H.C. Lee.
The research was supported by the National Institutes of Health and the US Department of Energy.