304 North Cardinal St.
Dorchester Center, MA 02124
With the biggest chipmakers for data centers keeping Russia out of next-generation devices and mobile and software companies leaving that market, it’s not surprising that Russian researchers are working quickly to find ways around the new technologies that will drive the rest of the world.
This is important for Russia right now, but it’s likely that these efforts will lead to similar ones in China, which is also used to implement tech sanctions, as we saw with Huawei, for example.
After Russia invaded Ukraine, the US stopped sending key tech products like semiconductors to Russia last week. At least AMD, Intel, TSMC, and GlobalFoundries have stopped sending chips to Russia because they want to follow US export rules. Oracle and SAP stopped doing business in the country last night, and Dell, HP, and Lenovo have also stopped sending products there.
Since key Russian banks were kicked out of the system on Wednesday, police have been keeping an eye on the Swiss data center that the financial messaging service SWIFT uses.
These sanctions have far-reaching consequences, affecting everything from Russia’s most powerful supercomputers to enterprise systems and the wider mobile world for both businesses and consumers.
” Professor Andrei Sukhov, who runs the CAD lab at Moscow’s HSE University, says, “In the meantime, it’s not enough to just say what the problem is. You need to use what you have to find a quick way out of the situation.
In a timely article for the Association for Computing Machinery (ACM), Sukhov talks about how Russian computer science teams are looking into building the next generation of clusters using older clustering technologies and a lot of open-source software for managing everything from code portability to parallelization, as well as standards like PCIe 3.0 and USB 4, and even existing Russian knock-off buses inspired by Infiniband (Angara ES8430).
Such systems would have to use processing and networking tools that are easy to find and use. If it comes down to it, Russia and China have better mobile options because they were made there.
Russia has seven supercomputers on the list of the 500 most powerful computers in the world. The computer with the most nodes in Russia is the 199-node super Chervonenkis, which is ranked number 19. Chervonenkis is made up of processors from AMD called Epyc and accelerators from Nvidia called A100. Without important parts like the InfiniBand, an integrator can’t do much.
The 36th, 40th, and 43rd most powerful machines all have AMD processors and Nvidia accelerators, which makes sense.
Sukhov said, “These devices aren’t the most powerful, but their production is completely independent and doesn’t affect the patent rights of American or European companies.”
Even if all the parts are there, you still need to make new boards. Sukhov said that this problem could be solved by switching between processors using wireless protocols, even though this would be bad for network latency. This would make it hard to do real HPC simulations with tight connections and low delays (which come in handy in areas like nuclear weapons simulations, as just one example).
“Given that available mobile systems-on-chip are on the order of 100 Gflops, small clusters of high-performance systems-on-chip can easily reach the performance of several teraflops,” said Sukhov.
“Using standard open operating systems like Linux will make it much easier to use custom applications and make it possible for these kinds of systems to run in the near future.” It is possible for these clusters to be different, with different systems-on-chip for different tasks (or, for example, FPGAs to make specialized accelerators for specific tasks that can be changed on the fly).
After the article came out, he had a short talk with The Register and said, “The supercomputers that are already in use are not expected to cause any problems.” Based on Linux, these supercomputers can keep running without help from the companies that made the hardware and software. So far as I know, all scientific supercomputers are run in normal mode today, even ones that are more than five years old. They can only be stopped by hacker attacks or by forcing them to do something. But I’ve never heard of anything like that happening with scientific projects like supercomputers.
“Russia can’t make a new supercomputer in the next few years, of course. Still, the way we’ve suggested can be used to meet all the current needs in computing and data processing. “Especially if we use different types of hardware acceleration for different tasks,” he says.
“It’s important to remember that our proposed method is meant to be put into action quickly as a pilot project.” During this implementation, software solutions, new protocols for exchanging data, and new computer technologies will be worked out.
“The cluster device will be able to be improved in the future.” A new motherboard with multiple chips connected by a bus, for example, could be created.”
READ MORE ARTICLES: