News on China's scientific and technological development.

antiterror13

Brigadier
Iron Delivery Man from JingDong!

Please, Log in or Register to view URLs content!


90UyVBU.jpg


In3iAMH.jpg

What is the purpose of the device at the back ?
 

subotai1

Junior Member
Registered Member
Not sure. If I have to guess, ir might be some kind of device that might protect the lower back of the workers who have to a lot of heavy lifting?
Its definitely a small exoskeleton with a couple support rods that go from the upper thighs to the strap and support above the waste. The goal appears to be to save the back and help power the lift when the worker picks up an object.
 

Hendrik_2000

Lieutenant General
Henri K take on supercomputer race
Please, Log in or Register to view URLs content!


If the supercomputer race was even tighter than ever before between the US and China
Please, Log in or Register to view URLs content!
, at the 49th TOP500 that took place in June in Germany this year, the gap between these two rivals are growing "suddenly" in
Please, Log in or Register to view URLs content!
.

Now, more than 40% of the fastest supercomputers in the world are located in China, which also jealously keeps the first two places on the podium for two consecutive years now. It will be noted that 15 years ago, no supercomputer of Chinese origin was on the list.

The United States, meanwhile, have only 143 machines classified in the new TOP500, against 169 before (and 160 for China), the lowest level ever since the creation of the ranking in 1993, but they still keep a very comfortable ahead of other countries, far ahead of Japan, 3rd in the world, with only 35.

Please, Log in or Register to view URLs content!

The 5 most powerful supercomputers in the world, in the 50th TOP500 ranking

With a computing power of 93 peta-flops per second, that is 93 million billion, or 93 × 10 15 , of floating-point operation per second, the supercomputer Sunway TaihuLight has beenmonopolizing the throne of TOP500 for four editions and still remains almost three times faster than the second ranking, another Chinese supercomputer Tianhe -2, which develops a power of 33 peta-flops.

The Sunway TaihuLight is also the world's first supercomputer to cross the bar of 100 peta-flops, more than 125 peta-flops at the peak, with Sunway's fully-featured 64-bit RISC processors in China. It is hosted today at the Wuxi National Supercomputing Center and co-operated by Tsinghua University in Beijing.

Third place on the podium goes to Switzerland's Piz Daint , the most powerful in Europe with just under 20 peta-flops on fire.

The "little" surprise of this new TOP500 is the big leap forward in terms of the number of Chinese supercomputers entered in the list. If China still had only 167 supercomputers classified exactly one year ago and 160 in June this year, the country now has 202 very high performance machines and now leaves the United States, with 143 machines, far behind .

On the other hand, with a fleet of 40.4% supercomputers in the TOP500, Chinese machines account for only 35.4% of total computing capacity. These figures are 28.6% and 29.6% for the United States, then 7% and 10.8% for Japan, the third most populous "supercomputer" country in the world.

This is reflected by the fact that few Chinese supercomputers are among the top 100 - 6 in total - which occupy respectively 1st, 2nd, then 52nd, 64th, 73rd and 96th place, against the presence of 33 US supercomputers.

But if we have observed a relative rarity and Chinese disparity in the first 100 world machines, their supercomputers take place equitably in all subsequent tranches.

For example, in the world's 100th to 199th supercomputer quadrant, 49 of them are Chinese. There are 58 for the 200th, 48 for the 300th, and 41 for the 400th.

Such an observation is interesting in the sense that this could demonstrate, in a way, that the development of supercomputer in China is not only dedicated to a race for more flops, but organized in a homogeneous and omnidirectional way.

This would make it possible to address requirements of different requirements in computing power, and thus avoid over or double investment and therefore waste in the actual operation, always having at the disposal of some ultra-powerful supercomputers at the forefront. of the spear for very complex calculations.

Please, Log in or Register to view URLs content!

Number of supercomputers classified in TOP50 by country (Image: East Pendulum)

Please, Log in or Register to view URLs content!

Total computing power by country (Image: East Pendulum)

Apart from this ratio of the number of machines to total capacity, it should be noted that Chinese supercomputers are, on average, less "green" than their competitors.

Although the data is not complete, but the energy efficiency of Chinese machines, which is measured by the amount of operations per second that can generate with 1 Watt of power consumed, is at a low enough level, only 1, 40 GFlops / W, far behind the most economical of all, the Japanese machines, with an average energy efficiency of 5.59 GFlops / W, followed by the Swiss at 4.51 and the Canadians at 4.14.

But there are also exceptions - the Sunway TaihuLight for example is not only the most powerful supercomputer in the world but also the second most energy efficient in China, with a rate of 6.05 GFlops / W.

The most eco-friendly Chinese machine comes back to Era-AI designed by the Chinese manufacturer Sugon for academic purposes. For each Watt spent, the supercomputer is able to perform 8.60 GFlops in operation.

The fact that nearly half of Chinese supercomputers, 100 out of 202, did not provide figures for their consumptions could also have skewed the average, in one way or another.

Please, Log in or Register to view URLs content!

Average energy efficiency of supercomputers by country (Image: East Pendulum)

As for France, the new TOP500 now has 18 French supercomputers, one more than the middle of the year, which puts France in 5th place behind Germany (21) but now in front of the United Kingdom ( 15). It thus reduces the gap with Germany which had 28 supercomputers classified previously and also puts distance from England which had 17 in the middle of the year.

The most powerful French supercomputer is the Pangea , designed by the American company Hewlett-Packard for Total Exploration Production. It has more than 220,000 Xeon E5-2680 processors for a computing power of 5.3 peta-flops per second, and now ranks 21st worldwide, slightly down two places.

Total's supercomputer is closely monitored by Tera-1000-2 , the 23rd in the world, which is being developed this time by Frenchman Bull for the French Atomic Energy Commission (CEA). With 208,896 Intel Xeon Phi 7250 processors, this supercomputer dedicated to research displays 4.97 peta-flops per second on the meter, for an energy efficiency of 3.98 GFlops / W, much better than the Pangea with its 1, 27 GFlops / W.

Today, it is difficult to say how far China can still maintain its lead over the United States. The DoE, United States Department of Energy , is preparing to allocate more than 258 million US dollars to six US companies to develop the supercomputer of tomorrow power exa-Flops, machines 10 times more powerful than the Chinese Sunway TaihuLight today, by 2021.

The Chinese, meanwhile, have already included the development of exa-Flops supercomputer in the 13th five-year plan (2016-2020) of the country. Three Chinese entities, the Sugon Information Industry (中 科 曙光) supported by the Chinese Academy of Sciences (CAS), the National University of Defense Technology (NUDT) and the Institute 56 of the Chinese General Staff, have already received the authorization and the funding to launch each their own prototype.

The goal is to finalize the first prototypes by 2018-2019, and operational models by 2020-2021.

To be continued.
 
according to DefenseOne China and the CIA Are Competing to Fund Silicon Valley’s AI Startups
Please, Log in or Register to view URLs content!

The U.S. intelligence community is upping its early-stage investments in machine-learning companies — but Beijing is pouring in far more.

A trio of new investments in Silicon Valley machine-learning startups shows that the U.S. intelligence community is deeply interested in artificial intelligence. But China is investing even more in these kinds of U.S. companies, and that has experts and intelligence officials worried.

Founded to foster new technology for spies, the 17-year-old
Please, Log in or Register to view URLs content!
has also helped boost commercial products. (Its investment in a little company called Keyhole helped produce Google Maps.) Compared to a venture capitalist firm whose early-stage investments are intended to make some money and get out, the nonprofit’s angle is longer term, less venture, more strategic, according to Charlie Greenbacker, In-Q-Tel’s technical product leader in artificial intelligence, machine learning, natural language processing, analytics, and data science.

“Our model is to put a little bit of pressure at the right spot to influence a company to make sure it develops things that are useful to our customers,” said Greenbacker, who estimated that their investments in a given startup generally amount to about one of every 15 dollars the company has.

Greenbacker
Please, Log in or Register to view URLs content!
In-Q-Tel investments that underline the specific areas that the CIA and other spy agencies want to use AI for: image recognition, natural language processing, and predictive analytics.

The first is Orbital Insight, a company that analyzes satellite images to discover trends and patterns on a global scale. As Orbital’s CEO James Crawford
Please, Log in or Register to view URLs content!
last year, “The number of ships docked at a Malaysian port, even the color of a wheat field in western Nebraska, are actually signs…visible indicators of economic activity, not just for a local region but for an entire global industry.”

Natural language processing is another a big area of AI investment. Greenbacker highlighted a company called
Please, Log in or Register to view URLs content!
, which sells a tool that can read and summarize text, useful for the CIA’s Directorate of Intelligence (now Directorate of Analysis) and the decidedly unsexy work of combing through the world’s court cases, patent filings, and newspapers for relevant data. Primer is the work of Sean Gourley, a co-creator of the
Please, Log in or Register to view URLs content!
analysis and visualization engine and one of the more celebrated young data scientists working at the intersection of Silicon Valley and national security.

What does a spy agency do with all of this data and analysis? Try to predict the future, of course, and the intelligence community has a big interest in AI that will help it succeed there as well. In-Q-Tel invested a company called
Please, Log in or Register to view URLs content!
, which says its predictive analytics engine can help retailers anticipate demand for certain products — and perhaps help intel analysts anticipate national-security problems. When the researchers
Please, Log in or Register to view URLs content!
the engine
Please, Log in or Register to view URLs content!
of published news stories, it predicted, with high confidence, a steady increase in negative news about the security situation in Ukraine in the early spring of 2014. And indeed, that happened. (The experiment was done retroactively, after the Russian invasion, when outcomes were known to the researchers — but not to the AI.)

Unlike other technological areas, today’s AI startups don’t necessarily need intelligence-community money, says Greenbacker: “There’s a lot more money flowing into the AI space than is typical” in the startup ecosystem.

“But putting in a little leverage in the right spot can bend all this toward the things we care about,” he said. “At the end of the day, it means the government doesn’t have to foot all of the bill to develop technology it cares about.”

Competing with China

The intelligence community’s investment in the valley’s AI startups lags foreign investors and China. And that has some folks very concerned.

“The entire government spent $1.1 billion on unclassified AI programs in 2015. The estimate for 2016 was $1.2 billion,” said Greenbacker. “Meanwhile, Softbank [a Japanese multinational conglomerate] has a $100-billion-dollar fund for this. The Chinese government in their most recent
Please, Log in or Register to view URLs content!
has put $150 billion in this.”

Last week, three U.S. senators proposed legislation to improve how the U.S. approves the sale or transfer of tech with national-security importance: U.S. Sens. John Cornyn, R-Texas; Dianne Feinstein, D-Calif.; and Richard Burr, R-N.C., the chairman of the Senate intelligence committee.

“By exploiting gaps in the existing CFIUS [Committee on Foreign Investment in the United States] review process, potential adversaries, such as China, have been effectively degrading our country’s military technological edge by acquiring, and otherwise investing in, U.S. companies,” Sen. Cornyn said
Please, Log in or Register to view URLs content!
about the Foreign Investment Risk Review Modernization Act, or FIRRMA. “This undermines our national security and highlights the imperative of modernizing the CFIUS review process to address 21st-century threats. This bill takes a measured approach by providing long overdue reforms to better protect our country, while also working to ensure that beneficial foreign investment is not chilled.”

In
Please, Log in or Register to view URLs content!
over the summer, Cornyn singled out AI, as well as the supply chains for microelectronics, as particularly deserving of extra scrutiny. Former Defense Department and White House officials were particularly skeptical that prohibiting the Chinese from investing in start-ups was the best method to tackle the problem. Some suggested instead that the U.S. government should simply try to outspend China.

“I have a lot of opinions about it. They aren’t always approved for public release,” said Greenbacker. He added that the size of foreign investment moving into the AI space in Silicon Valley “will catch the attention of smaller companies and researchers seeking funding.”

Matthijs Broer, the chief technical officer of the CIA’s Science and Technology directorate, also declined to comment on specific legislation. But Broer did say that far greater foreign investment in areas like artificial intelligence, including in U.S. startups, was “absolutely” an issue of vital importance.

“You bet they are national security issues. Of course that is under review on the U.S. government side,” he said at the Defense One Summit on Thursday.
 

taxiya

Brigadier
Registered Member
Its definitely a small exoskeleton with a couple support rods that go from the upper thighs to the strap and support above the waste. The goal appears to be to save the back and help power the lift when the worker picks up an object.
Your answer is the closest.
Here is a video in action.
Please, Log in or Register to view URLs content!


It is to support the waist by reducing the load. It does not help the arms to lift though.
 

tidalwave

Senior Member
Registered Member
China IC firm sets milestone with rollout of 3D NAND flash
Josephine Lien, Taipei; Willis Ke, DIGITIMES [Wednesday 15 November 2017]
Yangtze River Storage Technology (YMTC) under the Tsinghua Unigroup has successfully developed 32-layer 3D NAND flash chips, marking a new milestone and major technological breakthrough in China's memory-chip industry, according to industry sources.

The sources said that the successful development of 32-layer 3D NAND flash chips is expected to drive YMTC to step up development of 64-layer technology, standing a chance of breaking the dominance of the 3D NAND market by Samsung, SK Hynix, Micron/Intel, and Toshiba/Western Digital camps in the near future.

YMTC belongs to one of China's three major memory camps and the country's only player in the 3D NAND sector. The other two camps, Fujian Jinhua Integrated Circuit affiliated with Taiwan-based United Microelectronics (UMC) and Hefei Ruili Integrated Circuit teaming up with GigaDevice Semiconductor, focus on the DRAM sector.

YMTC was originally scheduled to launch its first 32-layer 3D NAND sample chips by the end of December, but the launch was advanced significantly as it smoothly put the chips in SSD (solid state disk) and successfully finished the test of applying the chips to terminal devices in the first half of November, the industry sources said.

The company has completed equipment installation and is slated to kick off trial production in the second quarter of 2018, with initial capacity set at 5,000 pieces per month. It will start mass production after achieving stable yield rates, the sources said.

The technical breakthrough has sent shockwaves through the semiconductor sector, as YMTC started to develop 3D NAND technology less than two years ago. Founded in July 2016, YMTC immediately acquired Wuhan Xinxin Semiconductor, a maker of NOR flash and SLC-type NAND memory with licensed technical support from US-based Spansion, and then moved to develop the advanced technology.

As 32-layer 3D NAND is just a groundwork for advancing to the 64-layer 3D NAND technology, YMTC is expected to step up its development of the 64-layer technology to narrow its technological gap with Korea's Samsung and SK Hynix as well as Micron, which are already developing 72-layer technology, the sources added.
 
Top