Tag Archives: #internet

Researchers Develop Speedier Network Analysis For a Range of Computer Hardware (Engineering / Computer Science)

The advance could boost recommendation algorithms and internet search.

Graphs — data structures that show the relationship among objects — are highly versatile. It’s easy to imagine a graph depicting a social media network’s web of connections. But graphs are also used in programs as diverse as content recommendation (what to watch next on Netflix?) and navigation (what’s the quickest route to the beach?). As Ajay Brahmakshatriya summarizes: “graphs are basically everywhere.”

Brahmakshatriya has developed software to more efficiently run graph applications on a wider range of computer hardware. The software extends GraphIt, a state-of-the-art graph programming language, to run on graphics processing units (GPUs), hardware that processes many data streams in parallel. The advance could accelerate graph analysis, especially for applications that benefit from a GPU’s parallelism, such as recommendation algorithms.

Brahmakshatriya, a PhD student in MIT’s Department of Electrical Engineering and Computer Science and the Computer Science and Artificial Intelligence Laboratory, will present the work at this month’s International Symposium on Code Generation and Optimization. Co-authors include Brahmakshatriya’s advisor, Professor Saman Amarasinghe, as well as Douglas T. Ross Career Development Assistant Professor of Software Technology Julian Shun, postdoc Changwan Hong, recent MIT PhD student Yunming Zhang PhD ’20 (now with Google), and Adobe Research’s Shoaib Kamil.

When programmers write code, they don’t talk directly to the computer hardware. The hardware itself operates in binary — 1s and 0s — while the coder writes in a structured, “high-level” language made up of words and symbols. Translating that high-level language into hardware-readable binary requires programs called compilers. “A compiler converts the code to a format that can run on the hardware,” says Brahmakshatriya. One such compiler, specially designed for graph analysis, is GraphIt.

The researchers developed GraphIt in 2018 to optimize the performance of graph-based algorithms regardless of the size and shape of the graph. GraphIt allows the user not only to input an algorithm, but also to schedule how that algorithm runs on the hardware. “The user can provide different options for the scheduling, until they figure out what works best for them,” says Brahmakshatriya. “GraphIt generates very specialized code tailored for each application to run as efficiently as possible.”

A number of startups and established tech firms alike have adopted GraphIt to aid their development of graph applications. But Brahmakshatriya says the first iteration of GraphIt had a shortcoming: It only runs on central processing units or CPUs, the type of processor in a typical laptop.

“Some algorithms are massively parallel,” says Brahmakshatriya, “meaning they can better utilize hardware like a GPU that has 10,000 cores for execution.” He notes that some types of graph analysis, including recommendation algorithms, require a high degree of parallelism. So Brahmakshatriya extended GraphIt to enable graph analysis to flourish on GPUs.

Brahmakshatriya’s team preserved the way GraphIt users input algorithms, but adapted the scheduling component for a wider array of hardware. “Our main design decision in extending GraphIt to GPUs was to keep the algorithm representation exactly the same,” says Brahmakshatriya. “Instead, we added a new scheduling language. So, the user can keep the same algorithms that they had before written before [for CPUs], and just change the scheduling input to get the GPU code.”

This new, optimized scheduling for GPUs gives a boost to graph algorithms that require high parallelism — including recommendation algorithms or internet search functions that sift through millions of websites simultaneously. To confirm the efficacy of GraphIt’s new extension, the team ran 90 experiments pitting GraphIt’s runtime against other state-of-the-art graph compilers on GPUs. The experiments included a range of algorithms and graph types, from road networks to social networks. GraphIt ran fastest in 65 of the 90 cases and was close behind the leading algorithm in the rest of the trials, demonstrating both its speed and versatility.

GraphIt “advances the field by attaining performance and productivity simultaneously,” says Adrian Sampson, a computer scientist at Cornell University who was not involved with the research. “Traditional ways of doing graph analysis have one or the other: Either you can write a simple algorithm with mediocre performance, or you can hire an expert to write an extremely fast implementation — but that kind of performance is rarely accessible to mere mortals. The GraphIt extension is the key to letting ordinary people write high-level, abstract algorithms and nonetheless getting expert-level performance out of GPUs.”

Sampson adds the advance could be particularly useful in rapidly changing fields: “An exciting domain like that is genomics, where algorithms are evolving so quickly that high-performance expert implementations can’t keep up with the rate of change. I’m excited for bioinformatics practitioners to get their hands on GraphIt to expand the kinds of genomic analyses they’re capable of.”

Brahmakshatriya says the new GraphIt extension provides a meaningful advance in graph analysis, enabling users to go between CPUs and GPUs with state-of-the-art performance with ease. “The field these days is tooth-and-nail competition. There are new frameworks coming out every day,” He says. But he emphasizes that the payoff for even slight optimization is worth it. “Companies are spending millions of dollars each day to run graph algorithms. Even if you make it run just 5 percent faster, you’re saving many thousands of dollars.”

This research was funded, in part, by the National Science Foundation, U.S. Department of Energy, the Applications Driving Architectures Center, and the Defense Advanced Research Projects Agency.

Featured image: MIT researchers developed software to more efficiently run graph applications on a range of computing hardware, including both CPUs and GPUs. Credits: Image: Istockphoto images edited by MIT News


Reference paper: “Compiling Graph Applications for GPUs with GraphIt”


Provided by MIT

Ultrathin Spray-applied MXene Antennas Are Ready for 5G (Material Science)

Drexel’s MXene technology reaches telecommunications performance benchmarks.

New antennas so thin that they can be sprayed into place are also robust enough to provide a strong signal at bandwidths that will be used by fifth-generation (5G) mobile devices. Performance results for the antennas, which are made from a new type of two-dimensional material called MXene, were recently reported by researchers at Drexel University and could have rammifications for mobile, wearable and connected “internet of things” technology.

Drexel University researchers have produced flexible, spray-applied antennas made from a two-dimensional material called MXene, that have reached performance measures on par with current telecommunications technology. ©Drexel University (Meikang Han)

The MXene antennas, which have been in development at Drexel for just over two years, are already performing nearly as well as the copper antennas found in most mobile devices on the market today, but with the benefit of being just a fraction of their thickness and weight.

“This combination of communications performance with extreme thinness, flexibility and durability sets a new standard for antenna technology,” said Yury Gogotsi, PhD, Distinguished University and Bach professor of Materials Science and Engineering in Drexel’s College of Engineering, who is the lead author of a paper on the MXene antennas recently published in the journal Advanced Materials. “While copper antennas have been the best in terms of performance for quite some time, their physical limitations have prevented connected and mobile technology from making the big leaps forward that many have predicted. Due to their unique set of characteristics MXene antennas could play an enabling role in the development of IoT technology.”

While mobile communications companies currently are on the cusp of introducing 5G technology, which could capitalize on an less-used portion of the telecommunication spectrum to enable faster data transmission, it will likely become the standard range of operation for new technology.

Beyond reaching performance capabilities, antennas for devices of the future must also be able to acquit themselves well in a variety of environments outside of the circuitboards of phones and computers. According to Gogotsi, this makes MXene an appealing material for new antennas because it can be spray applied, screen printed or inkjet-printed onto just about any substrate and remains flexible without sacrificing performance.

“Generally copper antenna arrays are manufactured by etching printed circuit boards, this is a difficult process to undertake on a flexible substrate,” said Meikang Han, PhD, a post-doctoral researcher at the A.J. Drexel Nanomaterials Institute who contributed to the research.”This puts MXene at a distinct advantage because it disperses in water to produce an ink, which can be sprayed or printed onto building walls or flexible substrates to create antennas.”

Ultrathin, spray-applied MXene antennas, developed by researchers at Drexel University, performs on par with antennas currently used in fifth-generation telecommunications technology. ©Drexel University (Meikang Han)

In the paper, Gogotsi and his collaborators, including Professor Gary Friedman, PhD, and Kapil Dandekar, PhD, E. Warren Colehower Chair Professor of the Electrical and Computer Engineering Department in Drexel’s College of Engineering, reported on the performance of three sets of spray-coated MXene antennas, which were between 7-14 times thinner and 15-30 times lighter than a similar copper antenna – even thinner than a coat of paint. They tested the antennas in both lab and open environments for key performance measures of how efficiently the antenna converts power into directed waves – gain, radiation efficiency and directivity. And they did the testing at the three radio frequencies commonly used for telecommunication, including one in the target frequency of operation for 5G devices.

In each instance, the MXene antennas performed within 5% percent of copper antennas, with performance increasing with thickness of the antenna. The best performing MXene patch antenna, about one-seventh the thickness of standard copper antennas, was 99% as efficient as a copper antennas operating at 16.4 GHz frequency in an open environment. MXenes were also 98% as effective as their copper counterparts operating in the 5G bandwidth.

Their performance exceeded that of several other new materials being considered for antennas, including silver ink, carbon nanotubes and graphene. And, siginificantly, these performance numbers did not waiver when the MXene antennas were subjected to as many as 5,000 bending cycles – a mark of durability that far surpasses its peer materials.

“MXene’s scalability and environmental sustainability in manufacturing has been well estabilished, for this material to now achieve performance goals on pace with the best materials on the market today is certainly a significant development,” Gogotsi said. “As we continue to test various coating patterns and techniques while additionally optimizing the composition of MXene materials, I expect their performance to continue to improve.”

References: Meikang Han, Yuqiao Liu, Roman Rakhmanov, Christopher Israel, Md Abu Saleh Tajin, Gary Friedman, Vladimir Volman, Ahmad Hoorfar, Kapil R. Dandekar, Yury Gogotsi. Solution‐Processed Ti 3 C 2 T x MXene Antennas for Radio‐Frequency Communication. Advanced Materials, 2020; 2003225 DOI: 10.1002/adma.202003225

Provided by Drexel University

Scientists Developed New Design Principles For Spin Based Quantum Materials (Quantum)

Northwestern University materials scientists have developed new design principles that could help spur development of future quantum materials used to advance (IoT) devices and other resource-intensive technologies while limiting ecological damage.

A crystal structure (left) and a visual model of the spin helix (right). Credit: Northwestern University

The study marks an important step in Rondinelli’s efforts to create new materials that are non-volatile, energy efficient, and generate less heat—important aspects of future ultrafast, low-power electronics and quantum computers that can help meet the world’s growing demand for data.

Rather than certain classes of semiconductors using the electron’s charge in transistors to power computing, solid-state spin-based materials utilize the electron’s spin and have the potential to support low-energy memory devices. In particular, materials with a high-quality persistent spin texture (PST) can exhibit a long-lived persistent spin helix (PSH), which can be used to track or control the spin-based information in a transistor.

Although many spin-based materials already encode information using spins, that information can be corrupted as the spins propagate in the active portion of the transistor. The researchers’ novel PST protects that spin information in helix form, making it a potential platform where ultralow energy and ultrafast spin-based logic and memory devices operate.

The research team used quantum-mechanical models and computational methods to develop a framework to identify and assess the spin textures in a group of non-centrosymmetric crystalline materials. The ability to control and optimize the spin lifetimes and transport properties in these materials is vital to realizing the future of quantum microelectronic devices that operate with low energy consumption.

The limiting characteristic of spin-based computing is the difficulty in attaining both long-lived and fully controllable spins from conventional semiconductor and magnetic materials. Their study will help future theoretical and experimental efforts aimed at controlling spins in otherwise non-magnetic materials to meet future scaling and economic demands.

Rondinelli’s framework used microscopic effective models and group theory to identify three materials design criteria that would produce useful spin textures: carrier density, the number of electrons propagating through an effective magnetic field, Rashba anisotropy, the ratio between intrinsic spin-orbit coupling parameters of the materials, and momentum space occupation, the PST region active in the electronic band structure. These features were then assessed using quantum-mechanical simulations to discover high-performing PSHs in a range of oxide-based materials.

The researchers used these principles and numerical solutions to a series of differential spin-diffusion equations to assess the spin texture of each material and predict the spin lifetimes for the helix in the strong spin-orbit coupling limit. They also found they could adjust and improve the PST performance using atomic distortions at the picoscale. The group determined an optimal PST material, Sr3Hf2O7, which showed a substantially longer spin lifetime for the helix than in any previously reported material.

Their approach provides a unique chemistry-agnostic strategy to discover, identify, and assess symmetry-protected persistent spin textures in quantum materials using intrinsic and extrinsic criteria. They proposed a way to expand the number of space groups hosting a PST, which may serve as a reservoir from which to design future PST materials, and found yet another use for ferroelectric oxides—compounds with a spontaneous electrical polarization. Their work will also help guide experimental efforts aimed at implementing the materials in real device structures.

This study has been republished from Science daily.

References: Xue-Zeng Lu, James M. Rondinelli. Discovery Principles and Materials for Symmetry-Protected Persistent Spin Textures with Long Spin Lifetimes. Matter, 2020; DOI: 10.1016/j.matt.2020.08.028 link: https://www.cell.com/matter/fulltext/S2590-2385(20)30455-0?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS2590238520304550%3Fshowall%3Dtrue

The Streisand Effect Says Sensoring Information Will Probably Backfire (Psychology)

Remember the infamous Sony email leak of 2014? It happened because the government of North Korea heard that Sony was releasing “The Interview,” a comedy about North Korean dictator Kim Jong Un. Hackers from the country leaked Sony’s corporate emails and threatened to bomb theaters that showed the film. Sony eventually pulled the film from theaters but released it online, and many independent theaters picked it up and screened it. The irony is that the film was lousy — it was panned by The New York Times, and has a dismal 52 percent rating on Rotten Tomatoes. But because it was central to such a dramatic controversy, many, many more people watched the movie than would have otherwise. That’s the Streisand effect: the phenomenon by which something you try to cover up becomes even more visible than it was in the first place.

The effect is named after legendary singer and actor Barbra Streisand. In 2003, photographer Kenneth Adelman took more than 12,000 aerial photos of California’s eroding coastline and posted them to a photography site. Somehow, Streisand found that the photos included an image of her clifftop mansion in Malibu, and the performer sued. According to Bloomberg, she alleged “that the photo invaded her privacy, violated the state’s anti-paparazzi statute and tried to profit from her name. She sought damages of more than $10 million, which she generously offered to donate to charity.”

Before the lawsuit, the photo had only six downloads — two of which were by Streisand’s own lawyers. After the suit made headlines, 420,000 people visited the photo site to see the images in the first month alone. Streisand’s lawsuit was dismissed, she had to pay Adelman’s six-figure legal fees, and the photo of her estate has now been seen millions of times. Whoops.

The image of Streisand Malibu house that led to the naming of the effect. Credit: Kenneth and Adelman, California cost record project

Censorship has always had a tendency to backfire — just consider the long history of attempts to ban controversial books, many of which are now classics. But with the internet, that backfire has even more force behind it. The Streisand effect has ignited a popularity in internet piracy, spread unflattering pictures of Beyoncé, and inadvertently led to $220,000 donated to charity. In 2014, before Uber was a household name, taxi drivers in several European cities protested the rideshare app and caused an 850 percent increase in new Uber users.

If it’s not already obvious, censoring backfires because people love forbidden fruit. People are curious by nature and if you tell them they can’t do something, they’ll want to do it — especially on the internet. As Andy Greenberg put it in Forbes, “A Web user and his information are like a grizzly and her cub. Come between them, and you’re likely to get mauled.”


Search Engines Make You Feel Smarter Than You Are (Psychology)

You’re having drinks with friends, and the conversation turns to whether or not beer is vegan. You whip out your phone and make a visit to your old pal Google: sure enough, some beer is made with fish bladders. Mystery solved! Another friend wonders: why does beer get you drunk, anyway? That’s a good question. How confident are you that you could answer it without the internet? According to research, you’re probably pretty confident—and you really shouldn’t be.

In 2015, Yale University doctoral student Matthew Fisher published a study in the Journal of Experimental Psychology to determine how the ability to search the internet affected how much people thought they knew. He gave subjects a survey asking simple questions such as “Why are there phases of the moon?” and “How does a zipper work?” and let only half of them use a search engine to find the answers. Next, he presented them with similar, but unrelated questions (“How do tornadoes form?” or “What is gluten?”) and only asked them to rate how well they thought they could answer each. Sure enough, those allowed to search online for the answers to the first questions rated their ability higher than those who answered them without help.

To make sure this wasn’t a quirk of the way he performed the study, Fisher redid his experiment in a bunch of different ways. He had subjects rate their knowledge first, then had half use the internet, then had them rate their knowledge again—same result. He made sure they both went to the same website for the answer, but had half go straight there and half search for the website—those who searched still rated their knowledge higher than the other group did. He used different search engines, filtered the results so the answer wouldn’t show up, and asked questions that didn’t have answers online. In every instance, the internet searchers showed an inflated sense of their own knowledge.

WHAT’S GOING ON?

In essence, you don’t know what you don’t know. “People are unlikely to be able to explain their own shortcomings,” Fisher told NPR. With search engines, “we are not forced to face our own ignorance and ask for help; we can just look up the answer immediately.” The ease with which you can find an answer on the internet is the exact thing that makes you overconfident in your abilities: you found the answer in three seconds flat just now, so why wouldn’t you be able to do it again?

Of course, these days, being without the internet is a rare thing. Most of the time, it might not make a difference how much of your knowledge is in your head and how much is out in cyberspace. Still, knowing your weaknesses is better than overconfidence…isn’t it?


References: (1) Matthew Fisher, Mariel K. Goddu, and Frank C. Keil, “Searching for Explanations: How the Internet Inflates Estimates of
Internal Knowledge”, Journal of Experimental Psychology: General, 2015, Vol. 144, No. 3, 674 – 687, http://dx.doi.org/10.1037/xge0000070 (2) https://www.npr.org/sections/health-shots/2015/04/02/396810355/searching-online-may-make-you-think-youre-smarter-than-you-are (3) https://theuncoverreality.wordpress.com/2020/08/06/the-dunning-kruger-effect-is-why-incompetent-people-think-theyre-great-psychology-mind-and-body/