System Bits: June 25

Supercomputers around the world
At last week’s International Supercomputing Conference in Frankfurt, Germany, the 53rd biannual list of the Top500 of the most powerful computing systems in the world was released.

Broken out by countries of installation, China has 219 of the world’s 500 fastest supercomputers, compared with 116 in the United States. Ranking by percent of list flops, the U.S. leads with 38.4%, compared with China’s 29.9%. Oak Ridge National Laboratory’s Summit system and Lawrence Livermore National Laboratory’s Sierra supercomputer accounted for a combined 243.2 petaflops between them, or a full sixth of the list’s capacity.

This analysis notes that, for the first time ever, the total aggregate score was more than 1.5 exaflops.

Of China’s 219 supercomputing systems, 74 came from Lenovo Group, 71 from Inspur, and 63 from Sugon. The last vendor now has some problems emanating from – where else? – the U.S. The U.S. Department of Commerce just added Sugon to the “entity list,” where you will also find Huawei Technologies. Also now on that list are Higon, Chengdu Haiguang Integrated Circuit, Chengdu Haiguang Microelectronics Technology, and Wuxi Jiangnan Institute of Computing Technology, all of which are involved in high-performance computing technology.

Ranked by installed petaflops, Lenovo led the list, followed by IBM, Cray, Hewlett Packard Enterprise (which is acquiring Cray), and Sugon. IBM gets props for having built two of the top-ranked supercomputers, Summit and Sierra.

Imec’s progress in battery technology
Imec had some news last week at the European Electric Vehicle Batteries Summit in Berlin, Germany. The Belgian research and development institute announced that it has come up with a solid-state lithium-metal battery cell with an energy density of 400 Wh/liter at a charging speed of 0.5C (2 hours). Imec also announced that it has started to upscale the materials and processes in a pilot line for fabrication of solid-state pouch cells at the EnergyVille Campus in Genk, Belgium, and is set-up in collaboration with the University of Hasselt. With its engineering roadmap for solid-state batteries, imec aims to surpass wet lithium-ion battery performance and reach 1000Wh/L at 2-3C by 2024.

The solid nanocomposite electrolyte that the R&D center has developed has an exceptionally high conductivity of up to 10 mS/cm with a potential for even higher conductivities. A distinguishing feature of the new material is that it is applied as a liquid – via wet chemical coating – and only afterwards converted into a solid when it is already in place in the electrodes. That way it is perfectly suited to be casted into dense powder electrodes where it fills all cavities and makes maximum contact, just as a liquid electrolyte does.

Using that solid nanocomposite electrolyte in combination with a standard lithium iron phosphate (LFP) cathode and lithium metal anode, imec has now fabricated an improved battery with an energy density of 400 Wh/liter at a charging speed of 0.5C (2 hours), a record combination for a solid-state battery. With this result, imec managed to double its excellent results of last year, following its roadmap to eventually reach densities over 1,000Wh/liter at a charging speed of 2-3C (less than half an hour).

Imec is working not only on batteries for electric vehicles – it is also focusing on battery-powered wearable electronics. Several battery types are being considered.

It is investigating the potential of the solid-state Li-ion battery, the 3D thin-film battery for micro-storage of energy, the nano-composite electrolyte, and the powder-composite battery for large storage systems.

Lancaster’s claim to a universal memory chip
Researchers at Lancaster University in the United Kingdom report their development of an ultra-low-power memory chip design which they say could yield revolutionary improvements in computing technology.

The scientists say this memory chip design, when implemented, could reduce peak power consumption in data centers by 20%. The team has received a U.S. patent on the design, with another patent pending.

Lancaster asserts that this memory design could challenge the massive DRAM market, worth $100 billion a year, as the working memory in computers, as well as the long-term memory in flash-based drives.

For years, information technology scientists and engineers have dreamt about a “universal memory” that could fit into all data storage applications.

Physics Professor Manus Hayne of Lancaster University said: “Universal Memory, which has robustly stored data that is easily changed, is widely considered to be unfeasible, or even impossible, but this device demonstrates its contradictory properties.”

The inventors of the device used quantum mechanics to solve the dilemma of choosing between stable, long-term data storage and low-energy writing and erasing.

While writing data to DRAM is fast and low-energy, the data is volatile and must be continuously ‘refreshed’ to avoid it being lost: this is clearly inconvenient and inefficient. Flash stores data robustly, but writing and erasing is slow, energy intensive and deteriorates it, making it unsuitable for working memory.

Professor Hayne said, “The ideal is to combine the advantages of both without their drawbacks, and this is what we have demonstrated. Our device has an intrinsic data storage time that is predicted to exceed the age of the Universe, yet it can record or delete data using 100 times less energy than DRAM.”

The university reports that several companies have expressed an interest or are actively involved in the research.