Association for Computing Machinery
Welcome to the September 12, 2012 edition of ACM TechNews, providing timely information for IT professionals three times a week.

ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).

HEADLINES AT A GLANCE


Intel and HP to Build World's Most Efficient Supercomputer
Techworld (09/11/12) Curtis. Sophie

Researchers at Hewlett-Packard and Intel are developing an energy-efficient supercomputing system for the U.S. Department of Energy's National Renewable Energy Laboratory (NREL). The system will be powered by a combination of current 32nm Xeon E5 processors and future 22nm Ivy Bridge processors, together with about 600 Xeon Phi co-processors. When fully operational, the system's total peak performance should exceed one petaflop, making it the largest supercomputer dedicated solely to renewable energy and energy efficiency research. The installation will use warm water liquid cooling technology to maximize the reuse of heat. Excess heat from the system will be guided into neighboring offices and labs and sent to other areas of the NREL campus to reduce central heating costs. The cooling system should help the NREL facility to become the world's most efficient data center, with a power usage effectiveness rating of at least 1.06. "At NREL, we have taken a holistic approach to sustainable computing," says NREL's Steve Hammond. "This new system will allow NREL to increase our computational capabilities while being mindful of energy and water used." The system is scheduled for completion next summer.


Researchers Craft Program to Stop Cloud Computer Problems Before They Start
NCSU News (09/10/12) Matt Shipman

North Carolina State University (NCSU) researchers have developed software that prevents performance disruptions in cloud computing systems by automatically identifying and responding to abnormal activities before they develop into larger problems. The program analyzes the amount of memory being used, network traffic, central processing unit (CPU) usage, and other data in a cloud computing infrastructure to determine what behaviors can be classified as normal. The software defines normal behavior for every virtual machine in the cloud and then looks for changes that could affect the system's ability to provide service to its users. If the program identifies a virtual machine that is acting abnormally, it runs a diagnostic that can determine which metrics are affected without exposing other data. "If we can identify the initial deviation and launch an automatic response, we can not only prevent a major disturbance, but actually prevent the user from even experiencing any change in system performance," says NCSU professor Helen Gu. She notes that once the system is operational, it uses less than one percent of the CPU load and 16 megabytes of memory. During testing, the system identified 98 percent of anomalies.


Southampton Engineers a Raspberry Pi Supercomputer
University of Southampton (United Kingdom) (09/12/12)

University of Southampton researchers have developed Iridis-Pi, a supercomputer made from 64 Raspberry Pi computers and Lego. "We installed and built all of the necessary software on the Pi, starting from a standard Debian Wheezy system image, and we have published a guide so you can build your own supercomputer," says Southampton professor Simon Cox. Iridis-Pi runs off of one 13 Amp mains socket and uses Message Passing Interface to communicate between nodes using Ethernet. The researchers note the entire system cost less than 2,500 pounds Sterling and includes 64 processors and one terabyte of memory. The researchers used the Python Tools for Visual Studio plug-in to develop software for the system. "The team wants to see this low-cost system as a starting point to inspire and enable students to apply high-performance computing and data handling to tackle complex engineering and scientific challenges as part of our ongoing outreach activities," Cox says.


Who's Trustworthy? A Robot Can Help Teach Us
New York Times (09/10/12) Tara Parker-Pope

Researchers at Northeastern University, the Massachusetts Institute of Technology, and Cornell University say they have found specific behaviors that seem to warn the human brain that another person cannot be trusted. First, the researchers filmed students interacting with other students they had never met in a game designed to elicit untrustworthy behavior. The researchers found that cues such as leaning away from someone, crossing arms in a blocking fashion, rubbing the hands together, and touching oneself on the face were indicators of untrustworthy behavior. "The more you saw someone do this, the more intuition you had that they would be less trustworthy," says Northeastern professor David DeSteno. The researchers then set up the same experiment with students playing the game with a friendly-faced robot. Some of the robots did not perform the untrustworthy cues while others did, and the students rated them as more untrustworthy. "It makes no sense to ascribe intentions to a robot, but it appears we have certain postures and gestures that we interpret in certain ways," says Cornell professor Robert H. Frank. The study suggests there could be an evolutionary benefit to cooperation, and to being able to identify untrustworthy people.


Get Ready for Computers Worldwide to Automatically Smother Cyber Strikes
NextGov.com (09/10/12) Aliya Sternstein

The U.S. government is planning to build a cyberecosystem that would prompt computers around the world to instantly suppress cyberattacks. The ecosystem would take collective action to galvanize cooperation among networks, external devices, and consumers, according to the Obama administration. "Computer systems, devices, applications, and users will automatically work together in near real-time to anticipate and prevent cyberattacks, automatically respond to attacks while continuing normal operations, evolve to address new threats, limit the spread of attacks across participating devices," and share timely security information, says a government research solicitation. The Department of Homeland Security (DHS) and the National Institute of Standards and Technology (NIST) are currently seeking public input on the potential benefits and challenges of the plan. "This information will help DHS and NIST develop future cyberecosystem security capabilities and an implementation strategy that will strengthen the security of critical infrastructures, federal information systems, and the private sector," the solicitation says. However, the cyberecosystem would require worldwide acceptance from industry, governments, universities, and consumers, according to officials. The solicitation also aims to determine if the field of artificial intelligence can support the concept of the cyberecosystem.


Disaster Is Just a Click Away
Kansas State University News (09/11/12) Greg Tammen

Kansas State University scientists are researching how to help computer users who have little computer experience improve their Web browsing safety without security-specific education, with the goal of keeping users from making mistakes that could compromise their online security. The researchers are developing a visual-messaging system called Education-optional Security Usability on the Internet, which shows novice computer users an easily understandable warning describing their security decisions. The researchers say the goal is to enable users to make a good gut reaction decision based on the message. "We want people to make good choices without having to understand the technical detail, but we don't want to make the choice for them; we want to show them the importance and danger level of that choice," says Kansas State professor Eugene Vasserman. The researchers are developing, testing, and evaluating the effectiveness of new and existing educational tools to determine which ones lead to better online security choices. Vasserman says the system should minimize the use of traditional text warnings and icons. The researchers also are designing a secure network for hospitals so medical devices can communicate with each other to monitor and relay information about a patient's health.


Computer, Read My Lips
Phys.Org (09/10/12)

Manipal International University researchers have developed a system that can interpret human emotions based on lip patterns, using a genetic algorithm that can match irregular ellipse fitting equations to the shape of the human mouth displaying different emotions. The researchers say they have trained the computer to recognize happiness, sadness, fear, anger, disgust, and surprise, as well as a neutral expression, by using photos of individuals from Japan and Southeast Asia. The upper and lower lips are analyzed as two separate ellipses by the algorithm. "In recent years, there has been a growing interest in improving all aspects of interaction between humans and computers especially in the area of human emotion recognition by observing facial expression," the researchers note. Initial applications of the system could be in helping disabled patients to interact more effectively with computer-based communication devices, according to the researchers.


Fujitsu to Build Software Robot to Pass College Entrance Exams
IDG News Service (09/10/12) Jay Alabaster

Fujitsu will develop software designed to enable artificial intelligence to pass the math portion of the University of Tokyo's entrance exam by 2021. Fujitsu says the software will complete exactly the same math test as a teenage applicant, solving exam problems quickly and without mistakes. The project will need to process text and formulas meant for human eyes, extract math problems and convert them into a form meant for computers, and then solve the problems at the level of the best high school students. "Each of these steps still poses major theoretical and practical problems, and for each one, the solution will involve an appropriate combination of various technologies," Fujitsu says. Computers on their own are only able to solve about half of the problems on previous exams, even when using advanced algorithms. Fujitsu's effort is part of a larger project run by the National Institute of Informatics to pass all subjects of the top university's entrance exams. The technology developed by Fujitsu could potentially be used to build intelligent systems that are capable of performing advanced analysis on themselves, for self-optimization and similar tasks.


Computer IDs Culprits With Tattoo Recognition
InnovationNewsDaily (09/06/12) Francie Diep

University of Colorado at Colorado Springs researchers have developed image-recognition technology that could help law enforcement identify criminals by their tattoos and other body marks. The researchers wrote a program that examines the tattoos, scars, moles, and other skin markings in a photo and then searches for likely matches in a photo database. The program can find similar tattoos that are not exactly the same but could be used to identify gang members who get coordinating designs. The program also can produce matches based on eyewitness descriptions. "This paper describes work on something of an exotic topic, and introduces improvements that are meant to move past proof-of-concept toward more practical tools," says the University of Notre Dame's Kevin Bowyer. The Colorado researchers, led by professor Terrance Boult, also wrote an artificial intelligence program that learns from examples. The researchers gathered random photos from the Internet to teach the program to find and match tattoos by appearance. The machine-learning algorithm also can learn how humans describe tattoos. Although the program currently recognizes 15 words, the researchers hope to further develop the system to recognize more than 100 tattoo descriptors.


Towards Computing With Water Droplets
Aalto University (09/06/12)

Water droplets could serve as bits of digital information for a new type of computing, according to researchers at Aalto University. The team used superhydrophobic tracks to guide water droplets along designed paths, and discovered that upon collision with each other on the high water-repellent surface, the two droplets rebound as if they were billiard balls. The researchers built a memory device in which water droplets act as bits of digital information, and demonstrated devices for elementary Boolean logic operations. Moreover, when the water droplets are loaded with reactive chemical cargo, the onset of a chemical reaction could be controlled by droplet collisions. The combination of the collision-controlled chemical reactions with droplet logic operations could allow for programmable chemical reactions in which single droplets serve simultaneously as miniature reactors and bits for computing. The researchers say technology based on superhydrophobic droplet logic could be used for autonomous simple logic devices not requiring electricity, as well as programmable biochemical analysis devices.


Penn Researchers Make First All-optical Nanowire Switch
Penn News (09/07/12) Evan Lerner

University of Pennsylvania researchers say they have developed the first all-optical photonic switch using cadmium sulfide nanowires and combined them into a logic gate. The technology was based on earlier research, which showed that cadmium sulfide nanowires exhibited strong light-matter coupling, making them very efficient at manipulating light. The researchers say the biggest challenge for photonic structures on the nanoscale is getting the light in, manipulating it, and then getting it out. "Our major innovation was how we solved the first problem, in that it allowed us to use the nanowires themselves for an on-chip light source," says Pennsylvania professor Ritesh Agarwal. Their technique involves cutting a gap into a nanowire and inserting enough energy into the first segment so that light is emitted from its end and through the gap. The researchers were able to measure the intensity of the light coming out of the end to show that the switch could effectively represent the binary states used in logic devices. "We used these optical switches to construct a NAND gate, which is a fundamental building block of modern computer processing," says Pennsylvania researcher Brian Piccione.


Great Barrier Reef Sea Squirt Could Create a Smaller, Faster, Greener Computer of the Future
University of Aberdeen (09/07/12) Jennifer Phillips

Researchers at the University of Aberdeen and the University of St. Andrew have developed a type of computer chip using molecules from a sea squirt found at the bottom of the Great Barrier Reef. They say their research could lead to a new generation of computers that are greener to produce, process more information, and are smaller than existing machines. "This project is looking at a greener, more sustainable alternative [to silicon]--making transistors from single molecules sourced from nature," says Aberdeen professor Marcel Jaspars. The researchers are focusing on the molecule patellamide, which is taken from the sea squirt Lissoclinum patella. They have developed a method for redesigning and producing patellamide molecules for use as computer components using clean and green biological processes. Jaspars says the method is "greener to produce as we can essentially 'grow' the parts required for the new 'patellamide' computer chip in a test tube, meaning it would be significantly more environmentally-friendly than creating silicon computer parts." He says the research could result in a smaller, more compact computer as the chip would have an array of single patellamide molecules.


25 of Today's Coolest Network and Computing Research Projects
Network World (09/06/12) Bob Brown

Many university labs are working on research projects that could have lasting effects on computer and networking technologies. For example, Carnegie Mellon University researchers have developed Duolingo, a free Web site that helps users learn new languages while also translating the text on Web pages into different languages. Princeton University researchers have developed Serval, an open source system that aims to make Web services such as Gmail and Facebook more easily accessible, regardless of where the end user is. Technical University in Darmstadt researchers have developed a method for home Wi-Fi routers to form a backup mesh network to be used by the police, firefighters, and other emergency personnel in the case of a disaster or other incident that disrupts standard phone networks. University of Tulsa researchers want to slow down network traffic in order to detect malware more efficiently. University of Washington researchers have developed Control-Alt-Hack, a card game designed to introduce computer science students to security topics. Bonn University researchers have developed a virtual drive that runs inside a USB drive to capture malware before it infects the computer. And University of California, Berkeley researchers have developed the Internet Protocol (IP) over Xylophone Players system, which can provide a fully compliant IP connection between two computers.


Abstract News © Copyright 2012 INFORMATION, INC.
Powered by Information, Inc.


To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.
Non-Members: Unsubscribe