Although molecular dynamics (MD) simulations of biomolecular systems often run for days to months, many events of great scientific interest and pharmaceutical relevance occur on long time scales that remain beyond reach. We present several new algorithms and implementation techniques that significantly accelerate parallel MD simulations compared with current stateof-the-art codes. These include a novel parallel decomposition method and message-passing techniques that reduce communication requirements, as well as novel communication primitives that further reduce communication time. We have also developed numerical techniques that maintain high accuracy while using single precision computation in order to exploit processor-level vector instructions. These methods are embodied in a newly developed MD code called Desmond that achieves unprecedented simulation throughput and parallel scalability on commodity clusters. Our results suggest that Desmond's parallel performance substantially surpasses that of any previously described code. For example, on a standard benchmark, Desmond's performance on a conventional Opteron cluster with 2K processors slightly exceeded the reported performance of IBM's Blue Gene/L machine with 32K processors running its Blue Matter MD code.

No abstract

No abstract

This paper addresses cooperative search for multiple stationary ground targets by a group of unmanned aerial vehicles with limited sensing and communication capabilities. The whole surveillance region is partitioned into cells where each cell is associated with a probability of target existence within the cell, which constitutes a probability map for the whole region. Each agent keeps an individual probability map and updates the map individually with measurements according to Bayesian rule. A nonlinear transformation of the probability map is introduced to simplify the computation by linearizing the Bayesian update. A consensus-like distributed fusion scheme is proposed for multiagent map fusion. We prove that all the individual probability maps converge to the same one that reflects the true existence or nonexistence of targets within each cell. Coverage and topology control algorithms are designed for the path planning of mobile agents. Moreover, the performance of the fusion scheme for asynchronous implementations of sampling and communication is analyzed. Finally, the effectiveness of the proposed algorithms is illustrated via simulations.

Two-dimensional semiconductors can be used to build next-generation electronic devices with ultrascaled channel lengths. However, semiconductors need to be integrated with high-quality dielectrics—which are challenging to deposit. Here we show that single-crystal strontium titanate—a high-κ perovskite oxide—can be integrated with two-dimensional semiconductors using van der Waals forces. Strontium titanate thin films are grown on a sacrificial layer, lifted off and then transferred onto molybdenum disulfide and tungsten diselenide to make n-type and p-type transistors, respectively. The molybdenum disulfide transistors exhibit an on/off current ratio of 108 at a supply voltage of 1 V and a minimum subthreshold swing of 66 mV dec−1. We also show that the devices can be used to create low-power complementary metal–oxide–semiconductor inverter circuits.

Detecting the sources or destinations that have communicated with a large number of distinct destinations or sources during a small time interval is an important problem in network measurement and security. Previous detection approaches are not able to deliver the desired accuracy at high link speeds (10 to 40 Gbps). In this work, we propose two novel algorithms that provide accurate and efficient solutions to this problem. Their designs are based on the insight that sampling and data streaming are often suitable for capturing different and complementary regions of the information spectrum, and a close collaboration between them is an excellent way to recover the complete information. Our first solution builds on the standard hash-based flow sampling algorithm. Its main innovation is that the sampled traffic is further filtered by a data streaming module which allows for much higher sampling rate and hence much higher accuracy. Our second solution is more sophisticated but offers higher accuracy. It combines the power of data streaming in efficiently estimating quantities associated with a given identity, and the power of sampling in collecting a list of candidate identities. The performance of both solutions are evaluated using both mathematical analysis and trace-driven experiments on real-world Internet traffic.

No abstract

No abstract