System Bits: Sept. 20

Levitating nanoparticles; datacenter-on-chip; tight squeeze for electrons.


Improving Torque Sensing
In an advance that could bring new types of sensors and studies in quantum mechanics, Purdue University researchers have levitated a tiny nanodiamond particle with a laser in a vacuum chamber, using the technique for the first time to detect and measure its torsional vibration.

The team said the experiment represents a nanoscale version of the torsion balance used in the classic Cavendish experiment, performed in 1798 by British scientist Henry Cavendish, which determined Newton’s gravitational constant. A bar balancing two lead spheres at either end was suspended on a thin metal wire. Gravity acting on the two weights caused the wire and bar to twist, and this twisting – or torsion – was measured to calculate the gravitational force.

In the new experiment, an oblong-shaped nanodiamond levitated by a laser beam in a vacuum chamber served the same role as the bar, and the laser beam served the same role as the wire in Cavendish’s experiment.

This graphic represents a new experiment where levitating a nanodiamond with a laser in a vacuum chamber for the first time was used to detect and measure its “torsional vibration,” an advance that could bring new types of sensors and studies in quantum mechanics. (Source: Purdue University)

The change of the orientation of the nanodiamond caused the polarization of the laser beam to twist, the team reminded. And while torsion balances have played historic roles in the development of modern physics, now, an optically levitated ellipsoidal nanodiamond in a vacuum provides a new nanoscale torsion balance that will be many times more sensitive.

The detection of torsional vibration, which can be used for torque sensing and also to achieve torsional ground state cooling, could aid efforts to study quantum theory and realize potential applications in quantum information processing and high-precision measurement for sensors.

Researchers Target A New Paradigm For Big Data Computing

Carnegie Mellon researchers Diana Marculescu and Radu Marculescu have been awarded an NSF grant to develop a new paradigm for big data computing. Their project focuses on a datacenter-on-a-chip (DoC) design consisting of thousands of cores that can run compute- and data-intensive applications more efficiently compared to existing platforms.

From Data Centers to Wireless Datacenter on Chip (WiDoCs). (Source: Carnegie Mellon University Electrical and Computer Engineering)

The researchers reminded that datacenters and high performance computing clusters are dominated by power, thermal, and area constraints. They occupy large spaces and necessitate sophisticated cooling mechanisms to sustain the required performance levels. Their proposed DoC design consists of thousands of cores that communicate via a new communication infrastructure, while provisioning the system resources for the necessary power, performance, and thermal trade-offs.

Further, this approach lies squarely at the intersection of two major trends in integrated systems design: low power, and communication centric design.

Goals in the project are to design design a small-world wireless architecture as a communication backbone for many core-enabled Wireless Datacenter on Chip (WiDoC), while establishing physical layer design methods for highly-integrated 3-D WiDoC suitable for low latency data communication. The researchers also hope to evaluate latency-power-thermal trade-offs for the proposed WiDoC platform by considering relevant big data applications.

The proposed research brings together novel and interdisciplinary concepts from network-on-chip (NoC), wireless and complex networks, communication circuits, and optimization techniques aimed at single chip solutions for achieving data center-scale performance.

The researchers believe this work will help to establish an interdisciplinary research-based curriculum for high performance many-core system design meant to increase the number of students attracted to this area of engineering.

Diana Marculescu said, “Our research will impact numerous areas. Big data applications like social computing, life sciences, networking, and entertainment will benefit immensely from this new design paradigm that aims at achieving server-scale performance from hand-held devices.”

This is a joint project between Carnegie Mellon University and Washington State University.

Quantum Effects Observed In 1D Wires
In an advance that could be used to aid in the development of quantum technologies, including quantum computing, University of Cambridge researchers have observed quantum effects in electrons by squeezing them into one-dimensional ‘quantum wires,’ and observing the interactions between them.

Regime of a single 1D wire subband filled. (Source: University of Cambridge)

They said the ability to control electrons in this way may lay the groundwork for many technological advances, including quantum computers that can solve problems fundamentally intractable by modern electronics, the team explained. But before such technologies become practical researchers need to better understand quantum, or wave-like, particles, and more importantly, the interactions between them.

Squeezing electrons into a one-dimensional ‘quantum wire’ amplifies their quantum nature to the point that it can be seen, by measuring at what energy and wavelength (or momentum) electrons can be injected into the wire.

The Cambridge researchers have also tested the latest predictions of what should happen at high energies, where the original theory breaks down.

While theoretical activity in the past decade has led to new predictions of other ways of exciting waves among the electrons, as if a person entering a train pushes so hard some people fall over and knock into others much further down the carriage, these new ‘modes’ are weaker than the spin and charge waves and so are harder to detect.

As such, the collaborators of the Cambridge researchers from the University of Birmingham predicted that there would be a hierarchy of modes corresponding to the variety of ways in which the interactions can affect the quantum-mechanical particles, and the weaker modes should be strongest in very short wires.

Their results will be applied to better understand and control the behavior of electrons in the building blocks of a quantum computer.

Related Stories
Prior Week’s System Research Bits (Sept 13)
Faster parallel computing; breaking net neutrality deadlock; smartphones as design systems.