Entity

Time filter

Source Type


Zhang P.,State University of New York at Stony Brook | Liu L.,State University of New York at Stony Brook | Deng Y.,State University of New York at Stony Brook | Deng Y.,National Supercomputer Center in Jinan
Parallel Computing | Year: 2015

We present a new data-driven paradigm for solving mapping problems on parallel computers. This paradigm targets at mapping data modules, instead of task modules, onto multiple processing cores. By dependency analysis of data modules, we devise a data movement matrix to reduce the need of manipulating task program modules at the expenses of handling data modules. To visualize and quantify the complex maneuver, we adopt the parallel activities trace graphs introduced earlier. To demonstrate the procedure and algorithmic values of our paradigm, we test it on the Strassen matrix multiplication and Cholesky matrix inversion algorithms. Mapping tasks has been more widely studied while mapping data is a new approach that appears to be more efficient for data-intensive applications that are becoming prevalent for today's parallel computers with millions of cores. Source


Li W.,Institute of High Performance Computing of Singapore | Guo M.,National Supercomputer Center in Jinan | Zhang G.,Institute of High Performance Computing of Singapore | Zhang Y.-W.,Institute of High Performance Computing of Singapore
Physical Review B - Condensed Matter and Materials Physics | Year: 2014

MoS2, a member of transition metal dichalcogenides (TMDs), has recently emerged as an interesting two-dimensional material due to its unique mechanical, thermal, electronic and optical properties. Unlike graphene which possesses massless Dirac fermions with ultrahigh electron mobility, monolayer MoS2 is a direct band gap semiconductor. An interesting question arises: Can monolayer MoS2 also possess massless Dirac fermions with ultrahigh electron mobility? Here, using first-principles calculations, we show that a monolayer MoS2 allotrope, which consists of repeated square-octagon rings (abbreviated as so-MoS2 to distinguish it from the normal hexagonal lattice, h-MoS2) possesses both massless Dirac fermions and heavy fermions. Distinct from the p-orbital Dirac fermions of graphene, the Dirac fermions of so-MoS2 are d electrons and possess a Fermi velocity comparable to that of graphene. The Dirac cone structure in so-MoS2 demonstrated here greatly enriches our understanding on the physical properties of TMDs and opens up possibilities for developing high-performance electronic or spintronic devices. © 2014 American Physical Society. Source


Li W.,Institute of High Performance Computing of Singapore | Zhang G.,Institute of High Performance Computing of Singapore | Guo M.,National Supercomputer Center in Jinan | Zhang Y.-W.,Institute of High Performance Computing of Singapore
Nano Research | Year: 2014

Using density functional theory calculations, we have investigated the mechanical properties and strain effects on the electronic structure and transport properties of molybdenum disulfide (MoS2) nanotubes. At a similar diameter, an armchair nanotube has a higher Young's modulus and Poisson ratio than its zigzag counterpart due to the different orientations of Mo-S bond topologies. An increase in axial tensile strain leads to a progressive decrease in the band gap for both armchair and zigzag nanotubes. For armchair nanotube, however, there is a semiconductor-to-metal transition at the tensile strain of about 8%. For both armchair and zigzag nanotubes, the effective mass of a hole is uniformly larger than its electron counterpart, and is more sensitive to strain. Based on deformation potential theory, we have calculated the carrier mobilities of MoS2 nanotubes. It is found that the hole mobility is higher than its electron counterpart for armchair (6, 6) nanotube while the electron mobility is higher than its hole counterpart for zigzag (10, 0) nanotube. Our results highlight the tunable electronic properties of MoS2 nanotubes, promising for interesting applications in nanodevices, such as opto-electronics, photoluminescence, electronic switch and nanoscale strain sensor. [Figure not available: see fulltext.] © 2014 Tsinghua University Press and Springer-Verlag Berlin Heidelberg. Source


Ma X.,Shandong University | Dai Y.,Shandong University | Guo M.,National Supercomputer Center in Jinan | Huang B.,Shandong University
Journal of Physical Chemistry C | Year: 2013

The general low quantum efficiency of semiconductor-based photocatalysts significantly limits their large-scale application. Here, we reveal the potential role that surface distortion can play in enhancing the photocatalytic quantum efficiency as well as the underlying mechanism by using TiO2 as a model photocatalyst. Specifically, proper surface distortion in a {101} surface can significantly promote the participation of electrons in photocatalytic reactions and further facilitate the transfer of photogenerated electrons in the bulk region to this surface. Moreover, surface distortion also prevents the photogenerated holes from transferring to the surface layer, thus separating the photogenerated holes from electrons and reducing the high recombination rate of carriers, which is believed to result in the generally low photocatalytic activities of the {101} surface. For the {001} surface, the distorted surface greatly promotes the transfer of electrons from the subsurface atomic layer (the initial electron trapping sites) to the outermost atomic layer (where photocatalytic reactions generally occur) by eliminating the original energy barrier, and the trapping of electrons on surface Ti 5c dz2 orbital cannot only facilitate their participation in the photocatalytic reactions but also significantly reduce the carrier recombination rate in the surface region. The results presented here can be used to account for the experimental results that surface distortion in TiO 2 can substantially improve the quantum efficiency of its intrinsic absorption. © 2013 American Chemical Society. Source


Zhong Y.,CAS Institute of Computing Technology | Zhu X.,National Supercomputer Center in Jinan | Fang J.,CAS Institute of Computing Technology
Proceedings of the 1st ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data, BigSpatial 2012 | Year: 2012

Geospatial applications have become prevalent in both scientific research and industry. Spatio-Temporal query processing is a fundamental issue for driving geospatial applications. However, the state-of-the-art spatio-temporal query processing methods are facing significant challenges as the data expand and concurrent users increase. In this paper we present a novel spatio-temporal querying scheme to provide efficient query processing over big geospatial data. The scheme improves query efficiency from three facets. Firstly, taking geographic proximity and storage locality into consideration, we propose a geospatial data organization approach to achieve high aggregate I/O throughput, and design a distributed indexing framework for efficient pruning of the search space. Furthermore, we design an indexing plus MapReduce query processing architecture to improve data retrieval efficiency and query computation efficiency. In addition, we design distributed caching model to accelerate the access response of hotspot spatial objects. We evaluate the effectiveness of our scheme with comprehensive experiments using real datasets and application scenarios. Copyright © 2012 ACM. Source

Discover hidden collaborations