domingo, 12 de julho de 2015

Men may feel more threatened by female bosses, research finds

 

 

Men may feel threatened by female supervisors and act more assertively toward them than male bosses, which could disrupt the workplace with struggles over power dynamics, according to new research published by the Society for Personality and Social Psychology.

"The concept of masculinity is becoming more elusive in society as gender roles blur, with more women taking management positions and becoming the major breadwinners for their families," said lead researcher Ekaterina Netchaeva, an assistant professor of management and technology at Bocconi University in Milan, Italy. "Even men who support gender equality may see these advances as a threat to their masculinity, whether they consciously acknowledge it or not."

While women are underrepresented in senior management positions in the United States, they are almost on par with men at middle and lower management levels, according to Labor Department statistics. In three experiments, Netchaeva and her co-authors discovered that men feel more threatened when they answer to female bosses.

In an experiment with 76 college students (52 male, 24 female) at a U.S. university, participants were told they would negotiate their salary at a new job in a computer exercise with a male or female hiring manager. After the negotiation, participants took an implicit threat test where they guessed words that appeared on a computer screen for a fraction of a second. Participants who chose more threat-related words, including "fear" or "risk," were judged to feel more threatened.

Male participants who negotiated with a female manager exhibited more threat and pushed for a higher salary ($49,400 average), compared to men negotiating with a male manager ($42,870 average). The manager's gender didn't affect female participants, who negotiated for a lower salary ($41,346 average), reflecting a common trend where women tend to be less aggressive than men in negotiations, Netchaeva said.

In another experiment, 68 male college students had to decide how to split a $10,000 bonus with a male or female team member or supervisor. Male participants evenly split the money with male or female team members, but men felt more threatened by a female supervisor and tried to keep more money for themselves than with a male supervisor.

In a similar experiment conducted online with 370 adult participants (226 male, 144 female) from the United States, men were more receptive to female supervisors who were described as proactive and direct rather than self-promotive and power-seeking. Specifically, male participants tried to keep a larger share of the $10,000 bonus if the female manager was described as ambitious or power-seeking. Female participants offered roughly the same bonus amount to proactive or ambitious female managers.

Self-assertive behavior by men toward female bosses could disrupt the workplace dynamics, stifle team cohesiveness and negatively affect team performance, Netchaeva said.

"In an ideal world, men and organizations would be concerned by these findings and adjust their behavior accordingly. But if they don't, where does that leave women?" she said. "Given the strong societal norms surrounding masculinity, it may be difficult for men to recognize or change their behavior."

If men won't change their actions, then female supervisors may want to appear more proactive and less power-seeking to maintain smooth relationships in the workplace, Netchaeva said.

 

Role of microbiota in preventing allergies

 

 

A cluster of type 3 cells (shown in green) in a mouse colon. These cells are induced by the microbiota and block type 2 allergic reactions.

Credit: © Institut Pasteur

The human body is inhabited by billions of symbiotic bacteria, carrying a diversity that is unique to each individual. The microbiota is involved in many mechanisms, including digestion, vitamin synthesis and host defense. It is well established that a loss of bacterial symbionts promotes the development of allergies. Scientists at the Institut Pasteur have succeeded in explaining this phenomenon, and demonstrate how the microbiota acts on the balance of the immune system: the presence of microbes specifically blocks the immune cells responsible for triggering allergies. These results are published in Science on July 9, 2015.

The hygiene hypothesis suggests a link between the decline in infectious diseases and the increase in allergic diseases in industrialized countries. Improvements in hygiene levels necessarily lead to reduced contact with microbes that is paralleled by an increased incidence in allergic and autoimmune diseases, such as type 1 diabetes.

Epidemiological studies have substantiated this hypothesis, by showing that children living in contact with farm animals -- and therefore with more microbial agents -- develop fewer allergies during their lifetime. Conversely, experimental studies have shown that administering antibiotics to mice within the first days of life results in a loss of microbiota, and subsequently, in an increased incidence in allergy.

However, until now, the biological mechanisms underlying this phenomenon remained unclear. In this study published in Science, the team led by Gérard Eberl (head of the Microenvironment and Immunity Unit at the Institut Pasteur) shows that, in mice, symbiotic intestinal microbes act on the immune system by blocking allergic reactions.

Several types of immune response can be generated in order to defend the organism. The presence of bacterial or fungal microbes provokes a response from immune cells known as type 3 cells. These immune cells coordinate the phagocytosis and killing of the microbes. However, in the case of infection by pathogenic agents that are too large to be handled by type 3 cells (such as parasitic worms and certain allergens), the cells that organize the elimination of the pathogen, but also allergic reactions, are known as type 2 cells.

In this study, scientists at the Institut Pasteur have shown that type 3 cells activated during a microbial aggression act directly on type 2 cells and block their activity. Type 2 cells are consequently unable to generate allergic immune responses. This work demonstrates that the microbiota indirectly regulates type 2 immune responses by inducing type 3 cells.

These results explain how an imbalance in microbiota triggers an exaggerated type 2 immune response normally used to fight large parasites, but that also leads to allergic responses.

These findings represent an important milestone in understanding the balance between our various defense mechanisms. In terms of allergy treatment, a hitherto unexplored therapeutic approach consists therefore in stimulating type 3 cells by mimicking a microbial antigen in order to block allergy-causing type 2 cells.


Story Source:

The above post is reprinted from materials provided by Institut Pasteur. Note: Materials may be edited for content and length.


Journal Reference:

  1. C. Ohnmacht, J.-H. Park, S. Cording, J. B. Wing, K. Atarashi, Y. Obata, V. Gaboriau-Routhiau, R. Marques, S. Dulauroy, M. Fedoseeva, M. Busslinger, N. Cerf-Bensussan, I. G. Boneca, D. Voehringer, K. Hase, K. Honda, S. Sakaguchi, G. Eberl. The microbiota regulates type 2 immunity through ROR t T cells. Science, 2015; DOI: 10.1126/science.aac4263

Using garlic to combat antimicrobial resistant urinary tract infections

 

 

Garlic extract may be an effective weapon against multi-drug resistant strains of pathogenic bacteria associated with urinary tract infections (UTI), according to a recent study published in the Pertanika Journal of Tropical Agricultural Science.

Conducted by researchers at the Birla Institute of Technology and Sciences in India, the study found that "even crude extracts of [garlic] showed good activity against multidrug resistant strains where antibiotic therapy had limited or no effect. This provides hope for developing alternative drugs which may be of help in fighting the menace of growing antibacterial resistance," the team states.

Urinary tract infection is the second most common infectious disease encountered in community practice. Worldwide, about 150 million people are diagnosed each year with UTI, at a total treatment cost in the billions of dollars. Although UTI is usually treated with antibiotics, "emerging antimicrobial resistance compels us to look back into traditional medicines or herbal products, which may provide appropriate/acceptable alternative solutions," the authors argue.

Garlic (Allium sativum) has been traditionally used for the treatment of diseases since ancient times. A wide range of microorganisms -- including bacteria, fungi, protozoa and viruses -- are known to be sensitive to garlic preparations. Allicin and other sulphur compounds are thought to be the major antimicrobial factors in garlic.

In this study, the team found that 56% of 166 bacteria strains isolated from the urine of people with UTI showed a high degree of resistance to antibiotics. However, about 82% of the antibiotic resistant bacteria were susceptible to a crude aqueous extract of Allium sativum. According to the researchers, "ours is the first study to report the antibacterial activity of aqueous garlic extract against multidrug resistant bacterial isolates from infected urine samples leading to UTI."

"To conclude, there is evidence that garlic has potential in the treatment of UTI and maybe other microbial infections," says the team. "However, it is necessary to determine the bioavailability, side effects and pharmacokinetic properties in more detail."

For further information please see: http://www.pertanika.upm.edu.my/Pertanika%20PAPERS/JTAS%20Vol.%2038%20(2)%20May.%202015/09%20Page%20271-278%20(JTAS%200616-2014).pdf


Story Source:

The above post is reprinted from materials provided by Universiti Putra Malaysia (UPM). Note: Materials may be edited for content and length.


Graphene-based film has a thermal conductivity capacity that is four times that of copper.

 

 

This is graphene-based film on an electronic component with high heat intensity.

Credit: Johan Liu / Chalmers University of Technology

Researchers at Chalmers University of Technology have developed a method for efficiently cooling electronics using graphene-based film. The film has a thermal conductivity capacity that is four times that of copper. Moreover, the graphene film is attachable to electronic components made of silicon, which favours the film's performance compared to typical graphene characteristics shown in previous, similar experiments.

Electronic systems available today accumulate a great deal of heat, mostly due to the ever-increasing demand on functionality. Getting rid of excess heat in efficient ways is imperative to prolonging electronic lifespan, and would also lead to a considerable reduction in energy usage. According to an American study, approximately half the energy required to run computer servers, is used for cooling purposes alone.

A couple of years ago, a research team led by Johan Liu, professor at Chalmers University of Technology, were the first to show that graphene can have a cooling effect on silicon-based electronics. That was the starting point for researchers conducting research on the cooling of silicon-based electronics using graphene.

"But the methods that have been in place so far have presented the researchers with problems," Johan Liu says. "It has become evident that those methods cannot be used to rid electronic devices off great amounts of heat, because they have consisted only of a few layers of thermal conductive atoms. When you try to add more layers of graphene, another problem arises, a problem with adhesiveness. After having increased the amount of layers, the graphene no longer will adhere to the surface, since the adhesion is held together only by weak van der Waals bonds."

"We have now solved this problem by managing to create strong covalent bonds between the graphene film and the surface, which is an electronic component made of silicon," he continues.

The stronger bonds result from so-called functionalisation of the graphene, i.e. the addition of a property-altering molecule. Having tested several different additives, the Chalmers researchers concluded that an addition of (3-Aminopropyl) triethoxysilane (APTES) molecules has the most desired effect. When heated and put through hydrolysis, it creates so-called silane bonds between the graphene and the electronic component (see picture).

Moreover, functionalisation using silane coupling doubles the thermal conductivity of the graphene. The researchers have shown that the in-plane thermal conductivity of the graphene-based film, with 20 micrometer thickness, can reach a thermal conductivity value of 1600 W/mK, which is four times that of copper.

"Increased thermal capacity could lead to several new applications for graphene," says Johan Liu. "One example is the integration of graphene-based film into microelectronic devices and systems, such as highly efficient Light Emitting Diodes (LEDs), lasers and radio frequency components for cooling purposes. Graphene-based film could also pave the way for faster, smaller, more energy efficient, sustainable high power electronics."

The research was conducted in collaboration with Shanghai University in China, Ecole Centrale Paris and EM2C -- CNRS in France, and SHT Smart High Tech in Sweden.


Story Source:

The above post is reprinted from materials provided by Chalmers University of Technology. Note: Materials may be edited for content and length.


Journal Reference:

  1. Yong Zhang, Haoxue Han, Nan Wang, Pengtu Zhang, Yifeng Fu, Murali Murugesan, Michael Edwards, Kjell Jeppson, Sebastian Volz, Johan Liu. Improved Heat Spreading Performance of Functionalized Graphene in Microelectronic Device Application. Advanced Functional Materials, 2015; DOI: 10.1002/adfm.201500990

Norway could be Europe's 'green battery'

 

 

Norwegian hydropower could make Norway the "green battery" of Europe -- not by building new power plants, but by further developing the hydropower installations that were built out beginning at the turn of the last century.

The Hydraulic Laboratory at the Norwegian University of Science and Technology (NTNU) is filled with models of hydroelectric dams and water tunnels. Norway's steep and scenic Geiranger fjord is also on display in miniature. Mini mountain boulders sluice down into the fjord at high speed in order to research the tsunami effect on land when the Åknes mountain massif eventually breaks apart and lands in the bay.

There's also a brand new mini power plant, a 1:65 scale copy of Norway's Torpa hydropower plant. With 147 meters of piping, it is not really that mini. It is the world's first physical model of a waterway with an air-cushioned surge chamber -- a design that could be key to transforming Norway's hydropower network into an international resource, a green battery that could soak up Europe's excess wind and solar power and release it on demand. .

Doctoral student Kaspar Vereide designed the model and is studying new solutions for existing hydropower plants. The model provides answers to how the air-cushioned chamber can be optimally designed.

Improving power plants for more energy

Hydropower plant construction and research in Norway came to a near standstill after the last major expansion in the 1960s and 70s. The country had developed what it needed, and today 96 per cent of the electricity in Norway comes from the country's 937 hydropower stations. Indeed, Norway is the sixth largest hydropower producer in the world -- which is all the more amazing when you consider the entire country has just 5 million people.

However, Norway has the capacity to produce much more energy from hydropower and the potential to become Europe's "green battery." The idea behind a "green battery" is that excess power from Europe's growing network of solar arrays and wind farms could be sent to Norway to pump water up from lower reservoirs to higher reservoirs. Then, when Europe needs this power again, Norway just opens the tap and lets the water spin through its hydropower turbines.

The trick, is however, that Norway's hydropower plants were not necessarily designed for this on-and-off operation scheme. Thus, improving existing power plants, rather than developing new ones, is the focus of ongoing research.

The first step is to solve the challenges associated with increasing the capacity and flexibility of existing plants.

Vereide represents a new generation of hydropower researchers moving into Norway's so-called first energy revolution: electricity from water.

The battle between two geniuses

Electricity represents the modern era, and at the end of the 1800s two geniuses fought a fierce battle over the basis for future electrical technology. Thomas Alva Edison, inventor of the incandescent lamp, said the electrical era should be based on the principle of direct current (DC). Nikola Tesla, who worked with Edison, was just as sure that alternating current (AC) was the future. And both were equally convinced that they held the solution.

This disagreement between two intelligent and strong personalities led Tesla to split from Edison and establish his own firm, Tesla Electric Light & Manufacturing. Here he produced pioneering inventions, and among his most important ones were neon lights, radio technology, the induction motor, robotics, turbines for hydropower -- and AC. When he died he held 700 patents.

Tesla turned out to be right about which current the modern world would be based on. Through his inventions he laid the foundation for much of our modern technology, including AC.

Keeping the frequency stable

AC requires less energy to transport than DC, and it loses less energy during transport. AC is also less dangerous, and this was the main reason that AC became the choice at that time.

In direct current, the electrons all flow in the same direction, whereas in alternating current the electrons move back and forth. In the European electricity grid, they do this 50 times per second, which translates to power being generated at a frequency of 50 Hz (Hertz) in European countries.

"Electrical systems and equipment are made for this rate, so it's important that the rate remains stable at 50 Hz. You could destroy all the electronics in the country if the frequency in the Norwegian power grid were to go higher or lower. The hydropower plants control this frequency, since frequency is a result of the amount of power being produced and used at any given time," says Vereide.

"For example," he says, "if a smelting plant suddenly has to stop because a tree has fallen onto nearby power lines, you're immediately producing too much electricity compared to what's being used. This causes the electrical frequency to increase, and the challenge is to bring down production quickly enough."

Or in the opposite scenario: raising production quickly enough when electrical consumption increases, as when the rest of Europe suddenly needs the excess power they have sent to Norway to be stored.

An anthill of water tunnels

Water's inertia poses a problem, since speeding up or slowing down the water takes time. If it takes a long time to slow the water, it takes a correspondingly long time to reduce power production.

In Norway, most large power plants are built with very long power tunnels to transport water from the reservoir to the hydroelectric plant.

"Norwegian mountains are full of water tunnels. It's like an anthill. We have way more kilometres of water tunnels than we have road tunnels in this country," says Vereide.

Solution lies in the surge chamber

"The faster we can accelerate the water, the faster we can change the amount of generated current. And the solution lies in the surge chamber. First the water has to accelerate from the upper to the lower reservoir, and the length of that waterway can be many kilometres," says Vereide.

If the water can be temporarily stored close to the power station, this reduces the length of the waterway to be accelerated. The surge chamber provides just this interim storage. This can be built by blowing out a cavern inside the water tunnel, near the turbine where electricity is generated.

The caverns or surge chambers thus act as intermediate storage for the water and reduce the distance between the water reservoirs, speeding up the ability to modify water flow through the turbine.

"Surge chambers offer a solution -- but they also create a problem. This problem is called load fluctuation," Vereide says.

Risk of blowouts

If the plant is running at full blast and suddenly stops, the water level in the surge chamber increases. At the extreme, the water could overflow unmonitored into the top of the chamber, which normally has a ventilation tunnel.

And when the plant starts up, or turns up the speed of the water flow, the opposite happens -- the water level is drawn down. In the worst case, air could be sucked into the water tunnel, which could cause an uncontrolled blowout of air from the power plant.

"So we have to be able to control these load fluctuations that occur. Among other things, it's important to determine how big surge chambers need to be to function best. My task is to figure out the optimal design for surge chambers," says Vereide.

Europe's green battery

According to Vereide, this technology is highly relevant right now. Traditionally, plants have been run very smoothly and quietly, with few stops and starts to create these fluctuations. But a major expansion of existing hydropower plants is needed for Norway to become Europe's green battery in the future. The power plants will also need to be started and stopped much more often, and then the problem of load fluctuations increases significantly.

"We'll benefit a lot from developing new technologies, both in order to keep electrical frequency stable and to run power plants more aggressively to serve a large market. A lot of opportunities exist for further development, and this is what I'm researching," Vereide says.

Hunting for the golden formula

"One of the improvements I'm working on is called throttling. There is a steel cone built into the connection between the tunnel and the surge chamber, where you can reduce the tunnel diameter. The restriction reduces the amount of water that can actually flow into and out of the surge chamber, and that lessens load fluctuations.

Vereide elaborates: "A key element is to find the optimal throttle size. I've come a long way with it, but the throttle has to be adapted to the unique specs of each power plant. I'm working on finding a formula that can provide the solution for each power plant. My dream is to find this formula."

If he succeeds in figuring out the formula, he will have laid a golden egg for future hydropower industry.

New model for thermodynamics

A distinctively Norwegian technology is the use of air-cushioned surge chambers in hydropower plants. This requires that the surge chamber, which provides intermediate storage for the water and which helps to control the frequency, is blasted out deep inside the mountain. This air-cushioned chamber has no opening to the air above and therefore cannot get rid of excess water. Instead the chamber is filled with compressed air, which changes with the water level.

The air acts as a shock absorber does on cars and bicycles. When water flows into or out of the chamber, the air pressure dampens these fluctuations.

"As the water level moves up and down, a thermodynamic is created. That is, the air is compressed or it expands, and the air temperature changes. To better understand the physics of this, I've developed a theoretical model that describes the thermodynamics.

A scientific article about the model has been published in the Journal of Hydraulic Engineering.

The better understood the thermodynamics inside the air-cushioned chamber is, the more control can be had over flow dynamics and the frequency of the electrical network, which is essential at increased power production. Vereide is using the model to increase knowledge about the physical properties and design of air-cushioned surge chambers.

"So far, research has only a limited understanding of surge chamber thermodynamics. We haven't yet been able to quantify the transport of energy (heat) that occurs when the air is compressed in the air-cushioned chamber. This knowledge is important for an optimal design of air-cushioned chambers," says Vereide.


Story Source:

The above post is reprinted from materials provided by The Norwegian University of Science and Technology (NTNU). The original item was written by Idun Haugan. Note: Materials may be edited for content and length.


Journal Reference:

  1. Vereide, K., Tekle, T., and Nielsen, T. Thermodynamic Behavior and Heat Transfer in Closed Surge Tanks for Hydropower Plants. J. Hydraul. Eng., 2015, 141(6), 06015002 (in pres) DOI: 10.1061/(ASCE)HY.1943-7900.0000995

Mobile connectivity indoors has just got better

 

 

Researchers at IMDEA Networks invent the simplest solution available today to swiftly build a mobile wireless positioning system in a new indoor environment. Unlike other systems, it requires neither manual and costly offline pre-calibration nor any special hardware.

IMDEA Networks Institute Institute announces the successful completion of the research project "Opportunistic Timing Signals for Pervasive Mobile Location." It has accomplished remarkable scientific advances in the optimization of pervasive mobile location services, bringing nearer to hand the evolution of mobile wireless connectivity towards a seamless integration of navigation and network communications. Dr. Domenico Giustiniano, a Research Assistant Professor at IMDEA Networks, a Madrid-based research institute, has led the scientific work, which was financed by the Swiss Confederation represented by Armasuisse -- Science and Technology.

Despite the increasing interest in the area of mobile indoor localization, the positioning capability of Wireless Local Area Network (WLAN)-based mobile devices does not meet the joint goals of high accuracy and fast response time, given the constraint of using a commodity smartphone as target device.

To solve this, the project has built an approach that uses solely Time-of-Flight (ToF) measurements and relies on software upgrades of simple commercial off-the-shelf 802.11 chipsets that can be integrated in any access point (AP). This innovative solution not only reduces system implementation costs by using WiFi six dollar chipsets (used, for instance, by very famous Cisco APs) to collect and filter measurements, but also the novel filtering technique applied requires just a few samples to estimate the distance range.

To ensure its working capability, the system has been tested across different and heterogeneous setups and testbeds (including scenarios with strong indoor multipath), resulting in a median error of the distance of merely 1.7 − 2.4 m. Furthermore, it has also participated in indoor localization competitions achieving comparatively positive results. Amongst the advantages of the proposed solution are its cost-effectiveness (it runs on commodity WiFi hardware) and its ease of implementation (it is the simplest solution available today to swiftly build a positioning system in a new indoor environment). Unlike other systems, it requires neither manual and costly offline pre-calibration nor any special hardware.

A new research project called MATISSE has been launched to build on the success of IMDEA Networks' former research initiative in the area of Pervasive Mobile Localization. MATISSE will expand the scope of the prior project to include a new specialist research topic: Collaborative Wideband Spectrum Sensing Systems.

Knowledge about usage of the spectrum data and about the user's location is essential to build any communication protocol and service. With this in mind, MATISSE has a two-fold objective. On the one hand, it aims to improve current poor knowledge about usage of the electro-magnetic spectrum by introducing cyber-physical nodes that will be capable of monitoring the wideband spectrum at a very large scale, such as cities and nations. On the other hand, it will devise and build a pervasive localization system that is able to pin-point the position of a mobile device regardless of the environmental conditions.

The project "Opportunistic Timing Signals for Pervasive Mobile Location" operated from April 2014 to March 2015 and MATISSE was launched immediately after and until March 2016. Armassuise -- Science and Technology continues to provide funding to complete this work.


Story Source:

The above post is reprinted from materials provided by IMDEA Networks Institute. Note: Materials may be edited for content and length.


New network design exploits cheap, power-efficient flash memory without sacrificing speed

 

 

Random-access memory, or RAM, is where computers like to store the data they're working on. A processor can retrieve data from RAM tens of thousands of times more rapidly than it can from the computer's disk drive.

But in the age of big data, data sets are often much too large to fit in a single computer's RAM. The data describing a single human genome would take up the RAM of somewhere between 40 and 100 typical computers.

Flash memory -- the type of memory used by most portable devices -- could provide an alternative to conventional RAM for big-data applications. It's about a tenth as expensive, and it consumes about a tenth as much power.

The problem is that it's also a tenth as fast. But at the International Symposium on Computer Architecture in June, MIT researchers presented a new system that, for several common big-data applications, should make servers using flash memory as efficient as those using conventional RAM, while preserving their power and cost savings.

The researchers also presented experimental evidence showing that, if the servers executing a distributed computation have to go to disk for data even 5 percent of the time, their performance falls to a level that's comparable with flash, anyway.

In other words, even without the researchers' new techniques for accelerating data retrieval from flash memory, 40 servers with 10 terabytes' worth of RAM couldn't handle a 10.5-terabyte computation any better than 20 servers with 20 terabytes' worth of flash memory, which would consume only a fraction as much power.

"This is not a replacement for DRAM [dynamic RAM] or anything like that," says Arvind, the Johnson Professor of Computer Science and Engineering at MIT, whose group performed the new work. "But there may be many applications that can take advantage of this new style of architecture. Which companies recognize: Everybody's experimenting with different aspects of flash. We're just trying to establish another point in the design space."

Joining Arvind on the new paper are Sang Woo Jun and Ming Liu, MIT graduate students in computer science and engineering and joint first authors; their fellow grad student Shuotao Xu; Sungjin Lee, a postdoc in Arvind's group; Myron King and Jamey Hicks, who did their PhDs with Arvind and were researchers at Quanta Computer when the new system was developed; and one of their colleagues from Quanta, John Ankcorn -- who is also an MIT alumnus.

Outsourced computation

The researchers were able to make a network of flash-based servers competitive with a network of RAM-based servers by moving a little computational power off of the servers and onto the chips that control the flash drives. By preprocessing some of the data on the flash drives before passing it back to the servers, those chips can make distributed computation much more efficient. And since the preprocessing algorithms are wired into the chips, they dispense with the computational overhead associated with running an operating system, maintaining a file system, and the like.

With hardware contributed by some of their sponsors -- Quanta, Samsung, and Xilinx -- the researchers built a prototype network of 20 servers. Each server was connected to a field-programmable gate array, or FPGA, a kind of chip that can be reprogrammed to mimic different types of electrical circuits. Each FPGA, in turn, was connected to two half-terabyte -- or 500-gigabyte -- flash chips and to the two FPGAs nearest it in the server rack.

Because the FPGAs were connected to each other, they created a very fast network that allowed any server to retrieve data from any flash drive. They also controlled the flash drives, which is no simple task: The controllers that come with modern commercial flash drives have as many as eight different processors and a gigabyte of working memory.

Finally, the FPGAs also executed the algorithms that preprocessed the data stored on the flash drives. The researchers tested three such algorithms, geared to three popular big-data applications. One is image search, or trying to find matches for a sample image in a huge database. Another is an implementation of Google's PageRank algorithm, which assesses the importance of different Web pages that meet the same search criteria. And the third is an application called Memcached, which big, database-driven websites use to store frequently accessed information.

Chameleon clusters

FPGAs are about one-tenth as fast as purpose-built chips with hardwired circuits, but they're much faster than central processing units using software to perform the same computations. Ordinarily, either they're used to prototype new designs, or they're used in niche products whose sales volumes are too small to warrant the high cost of manufacturing purpose-built chips.

But the MIT and Quanta researchers' design suggests a new use for FPGAs: A host of applications could benefit from accelerators like the three the researchers designed. And since FPGAs are reprogrammable, they could be loaded with different accelerators, depending on the application. That could lead to distributed processing systems that lose little versatility while providing major savings in energy and cost.


Story Source:

The above post is reprinted from materials provided by Massachusetts Institute of Technology. The original item was written by Larry Hardesty. Note: Materials may be edited for content and length.