Association for Computing Machinery
Welcome to the June 19, 2013 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).


Scientist Out to Break Amdahl's Law
IDG News Service (06/17/13) Agam Shah

Speaking at this week's International Supercomputing Conference, JA1/4lich Supercomputing Center director Thomas Lippert led a presentation challenging Amdahl's generalized law, which focuses on performance relative to parallel and serial computing, by exposing it to a new class of experiments in parallel computing. The mathematics of Amdahl's law assume there is a limit to parallel acceleration as long as certain factors remain constant. Lippert's presentation was based on experiments done as part of the DEEP Project, which is investigating highly parallel computing models that help boost the performance of supercomputers. "My team was building the very effective JUROPA system together with Bull, Partec, and Intel," a machine that is "ideal for highly complex problems that exhibit a lower concurrency, in general," Lippert said. However, Lippert's goal is not to challenge Amdahl's law. "It is simply obvious that we should adapt the right piece of hardware to the corresponding concurrency," because only this approach has the potential to be most energy efficient and performance oriented at the same time, he said.

U.S. and Russia Sign Pact to Create Communication Link on Cybersecurity
Washington Post (06/17/13) Ellen Nakashima

The United States and Russia have announced a first-of-its-kind agreement to use real-time communications about national security incidents to lower the risk of conflict in cyberspace. The two countries will use the U.S. Nuclear Risk Reduction Center to alert each other of cyberexercises that could be misperceived as attacks. In addition, a secure phone link will be established to enable U.S. and Russian cybersecurity officials to speak directly in the event of an incident, a working group will be formed, and processes will be created for technical information exchange between the U.S. Computer Emergency Readiness Team and its Russian counterpart. "We recognize that threats to or in the use of [computer technologies] include political-military and criminal threats, as well as threats of a terrorist nature, and are some of the most serious national and international security challenges we face in the 21st century,” says a joint statement from President Barack Obama and President Vladimir Putin. "We view this cooperation as essential to safeguarding the security of our countries." The exchange of technical data called for in the agreement would be stripped of personal identifying information and would include Internet protocol addresses that point to computer networks that host malicious activity.

Mapping a Room in a Snap
Swiss Federal Institute of Technology in Lausanne (06/18/13)

Researchers at the Swiss Federal Institute of Technology in Lausanne (EPFL) have developed an algorithm that makes it possible to measure the dimensions of a room using a few microphones and a snap of the fingers. The microphones pick up the direct sound from the source, as well as the echoes arriving from various walls. EPFL's Ivan Dokmanic says the algorithm compares the signal from each microphone and analyzes the lag times to calculate the distance between the microphones, as well as the distance from each microphone to the walls and the sound source. In addition, by analyzing each echo’s signal using "Euclidean distance matrices," the system can tell whether the echo is rebounding for the first or second time and determine the unique "signature" of each of the walls. "Architects could use this to design rooms--for example concert halls or auditoriums--based upon the specific acoustics they would like to create," Dokmanic says. He also notes the algorithm could be used in mobile devices to determine location information inside buildings, where GPS signals are weak.

NASA Issues Grand Challenge, Calls for Public, Scientific Help in Tracking Threatening Asteroids
Network World (06/18/13) Michael Cooney

The U.S. National Aeronautics and Space Administration (NASA) recently announced two programs that call on scientists and organizations around the world to help spot, track, and possibly alter the direction of asteroids headed for Earth. One of the programs is part of NASA's Grand Challenges series, in which it calls on the public to develop unique solutions to difficult problems. This large-scale project will focus on detecting and characterizing asteroids and learning how to deal with potential threats, according to NASA. The challenge will include a variety of partnerships with other governmental agencies, international partners, industry, academia, and citizen scientists. NASA also put out a request for information (RFI) that invites industry and potential partners to offer ideas on accomplishing NASA's goal to locate, redirect, and explore an asteroid. NASA is seeking information for system concepts and innovative approaches for the agency's Asteroid Program, which involves redirecting an asteroid and placing it near the moon for study. The RFI is specifically looking for input on asteroid observation, asteroid redirection systems, integrated-sensing systems, refinements of the Asteroid Redirect Mission, applications of satellite-servicing technology, asteroid deflection demonstrations, asteroid-capture systems, and crew systems for asteroid exploration.

Now You Can Build Google’s $1M Artificial Brain on the Cheap
Wired News (06/17/13) Daniela Hernandez

Stanford Artificial Intelligence Lab director Andrew Ng has released a paper aimed at making deep learning more accessible to researchers by showing how to make a neural network that costs about $20,000 using powerful but inexpensive graphics-processing units (GPUs). Deep learning relies on a combination of hardware and software to imitate the functioning of the human brain. Last year at Google, Ng built a $1-million computerized brain that detects cat videos on Youtube. Ng's system taught itself to find the videos using a 1-billion-connection network on 1,000 computers. However, Ng says some researchers wondered how they could make progress in deep learning without that level of funding. “I hope that the ability to scale up using much less expensive hardware opens up another avenue for everyone around the world,” Ng says. “That’s the reason I’m excited--you can now build a 1-billion-connection model with $20,000 worth of hardware. It opens up the world for researchers to improve the performance of speech recognition and computer vision.” Ng's research also paves the way for more powerful GPU-based applications in the future.

Mathematical Algorithms Cut Train Delays
CORDIS News (06/17/13)

New software is reducing delays and waiting times for commuters using railways in European countries such as Germany, the Netherlands, and Greece. The Algorithms for Robust and online Railway optimization: Improving the Validity and reliability of Large scale systems (ARRIVAL) project has developed software to adapt railway schedules in real time following unforeseen disruptions. The software uses algorithms and a new concept called recoverable robustness. The idea involves mathematics, theoretical modeling, and competitive analysis, and is a way of measuring the robustness and recoverability of plans. The ARRIVAL team created new models and methods with algorithms to help operators with delay management. The researchers also developed a central repository for the collection and exchange of real-world data. The ARRIVAL system was employed to draft a new timetable for the Dutch national railway system, which accommodates about 5,500 trains daily and is now viewed as one of Europe's most efficient railway networks. More countries plan to adopt the technology, which can be applied to other areas that require scheduling, such as road traffic navigation systems, industrial workflow systems, e-commerce, grid computing networks, and healthcare.

IT Thought Leaders Tackle Global STEM Issues in Meetup
eWeek (06/14/13) Chris Preimesberger

The United Nations (UN) and other sponsors will financially back AdviseHer, an online community that uses social media and other pipeline programs to advocate for women and girls in science, technology, engineering, and mathematics (STEM) education. AdviseHer will recruit women in STEM companies, retirees, and former technology company employees to advise young women who are interested in IT education. Funds also will be raised for startups founded by women. Participants in a recent innovation workshop, which took place onboard a British Airways jet en route from San Francisco to London, selected the concept and presented it to the UN ITU Committee at the recent Decide Now Act Summit. British Airways sponsored the initiative, called UnGrounded, which brought together Silicon Valley entrepreneurs, educators, venture capitalists, inventors, and journalists for a meetup focused on improving STEM education. Participants included key executives from IBM, Microsoft,, and IDEO. The workshop's backers say this could be the first of many UnGrounded summits in the sky. "The whole idea is to get ideas people together in one place, unencumbered by email or texting, and give them a deadline to try and think of solutions to problems," says British Airways' Simon Talling-Smith.

Breaking Down Brain Function
Texas Advanced Computing Center (06/12/13) Aaron Dubrow

University of Texas at Austin professor Russell Poldrack's research encompasses psychology, neuroscience, and computer science to understand how the human brain functions. Poldrack notes that functional magnetic resonance imaging (fMRI) and supercomputers increasingly are making it possible to perform intensive data analyses of the brain. "Researchers are very interested in what has come to be known as brain reading or decoding," he says. Poldrack and his colleagues have created OpenfMRI, a Web-based supercomputer-powered workflow that makes it possible for researchers to process, share, compare, and analyze bran scans from many different studies. The data comes primarily from four main partners, including Stanford University, Harvard University, the University of Colorado, and Washington University. "The OpenfMRI database, and related data-sharing efforts, will enable large sample studies with the necessary power to understand individual differences as they relate to basic neuroscience questions and illness," says Stanford professor Anthony Wagner. The goal of the project is to collect enough brain data to develop a bottom-up perspective on brain function, Poldrack notes. He and his team demonstrated that combining the studies in the OpenfMRI database enabled the construction of a tree-of-thought based on the similarity of brain-image scans.

Obama, the ‘Big Data’ President
Washington Post (06/14/13) Nancy Scola

The Obama administration, which strongly believes in the power of data to drive government decision-making, has initiated several big data projects. Last year, the White House launched the $200-million Big Data Research and Development Initiative, listing more than 85 examples of big data projects at a range of federal agencies. For example, CyberInfrastructure for Billions of Electronic Records (CI-BER) is a collaborative, global project aimed at sharing Earth observation data. The Defense Department also is spending about $250 million a year on big data research and development. “In the same way that past federal investments in information technology R&D led to dramatic advances in supercomputing and the creation of the Internet, the initiative we are launching today promises to transform our ability to use big data for scientific discovery, environmental and biomedical research, education, and national security,” says White House Office of Science and Technology Policy director John P. Holdren. However, analysts say incidents such as the recent National Security Agency data collection controversy highlight the need for increased debate on big data use and its potential implications for the relationship between the government and large tech firms and for the public in terms of privacy and understanding what rights it is giving away to the government.

Titan Completes Acceptance Testing
Oak Ridge National Laboratory (06/12/13) Morgan McCorkle

Oak Ridge National Laboratory's (ORNL) Titan supercomputer has completed acceptance testing, which ensures the functionality, performance, and stability of the machine. Titan, which ranked as the fastest supercomputer in the world in the November 2012 Top 500 list, is capable of more than 27,000 trillion calculations a second, or 27 petaflops. "The purpose of Titan's incredible power is to advance science, and the system has already shown its abilities on a range of important applications and has validated ORNL's decision to rely on GPU accelerators," says Oak Ridge Leadership Computing Facility project director Buddy Bland. For example, LAMMPS, a high-performance molecular dynamics application, has seen more than a sevenfold improvement on Titan over its performance on a CPU-only system. There also has been an almost fourfold boost in speed on Titan by two other codes: Denovo, which simulates neutron transport in nuclear reactors; and WL-LSMS, which models the statistical mechanics of magnetic materials. "The system was delivered on schedule and within budget, and it has clearly shown its value as a research tool," Bland says.

This Robot Wants to Beat You at Air Hockey
IEEE Spectrum (06/13/13) Jason Falconer

Chiba University researchers have developed an air-hockey robot that can change its strategy in response to the playing style of its human opponent. Building on the Namiki Lab's work in high-speed tracking, the team has designed a system that consists of an air-hockey table, a Barrett four-axis robotic arm, two high-speed cameras, and an external PC. The system tracks the puck and opponent's paddle, and position data from camera images are then processed by the external PC, which determines the robot's next move. The robot tracks the game at a very fast rate of 500 frames per second. The researchers designed the robot to observe the speed and position of the player's paddle in relation to the puck, and it uses this data to estimate whether its opponent is playing aggressively or defensively. In experiments, the robot detected playing behaviors, and forced players to change their strategy; players said this made the game more exciting. So that the game is entertaining for human players, the researchers programmed the robot with a three-layer control system. The first layer controls basic hardware motion, the second decides the device's short-term strategy, and the third ascertains a long-term strategy.

Berkeley Lab Study Finds Moving Select Computer Services to the Cloud Promises Significant Energy Savings
Lawrence Berkeley National Laboratory (06/11/13) Jon Bashor

Researchers at Lawrence Berkeley National Laboratory and Northwestern University have found that moving common software applications used by 86 million U.S. workers to the cloud could save enough energy to power Los Angeles for a year. The researchers focused on email, customer relationship management, and bundled productivity software. Moving these applications from local computer systems to centralized cloud services could cut information technology energy consumption by up to 87 percent. The researchers aim to develop a model that both scientists and the general public could use to analyze the energy and carbon impacts of cloud computing. "What we found overall is that by hosting services on the cloud as opposed to locally, the savings are pretty robust," says Northwestern's Eric Masanet. In developing the public-use model, the researchers used the best available and most credible data from a variety of sources, including their own past work. By providing an easy-to-use model, the project aims to facilitate more discussion of the energy-related aspects of cloud computing. "By studying the present day and cloud scenarios, you can see the net energy and carbon-footprint benefits at a range of scales," says Berkeley Lab researcher and principal project investigator Lavanya Ramakrishnan.

Illinois-Intel Partnership Leads to Prototype for Debugging Innovations
I2PC News (06/07/13) Tom Moone

A new process for parallel programming systems developed by researchers at the University of Illinois at Urbana-Champaign and Intel is designed to combat accidental bugs and malicious intrusions into software. The process, called QuickRec, creates steps for retracing to learn where something went wrong. "As you are running your program, when you detect a bug, you can use this to go back and trace the bug--how it came in," says Illinois professor Josep Torrelas. "If you see a security intrusion, you go back and see how it arose. It allows you to go back and see exactly how it got there." QuickRec is a prototype for a multicore Intel architecture record and replay system for multithreaded programs. The researchers say QuickRec could point the way toward the next level of innovation in performance monitoring and debugging support for processors, and it does not affect processing speed. "It is just a hardware device that you install to monitor the machine,” says Illinois professor Sam King. "It doesn’t slow the machine down."

Abstract News © Copyright 2013 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