Source author record

Nan Zhu

Nan Zhu appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

hep-ph Distributed, Parallel, and Cluster Computing

Catalog footprint

What is connected

4works

2topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2016arXiv

PFO: A Parallel Friendly High Performance System for Online Query and Update of Nearest Neighbors

Nearest Neighbor(s) search is the fundamental computational primitive to tackle massive dataset. Locality Sensitive Hashing (LSH) has been a bracing tool for Nearest Neighbor(s) search in high dimensional spaces. However, traditional LSH systems cannot be applied in online big data systems to handle a large volume of query/update requests, because most of the systems optimize the query efficiency with the assumption of infrequent updates and missing the parallel-friendly design. As a result, the state-of-the-art LSH systems cannot adapt the system response to the user behavior interactively. In this paper, we propose a new LSH system called PFO. It handles query/update requests in RAM and scales the system capacity by using flash memory. To achieve high streaming data throughput, PFO adopts a parallel-friendly indexing structure while preserving the distance between data points. Further, it accommodates inbound data in real-time and dispatches update requests intelligently to eliminate the cross-threads synchronization. We carried out extensive evaluations with large synthetic and standard benchmark datasets. Results demonstrate that PFO delivers shorter latency and offers scalable capacity compared with the existing LSH systems. PFO serves with higher throughput than the state-of-the-art LSH indexing structure when dealing with online query/update requests to nearest neighbors. Meanwhile, PFO returns neighbors with much better quality, thus being efficient to handle online big data applications, e.g. streaming recommendation system, interactive machine learning systems.

preprint2014arXiv

$B\to π\ell^{+}\ell^{-}$ decays revisited in the standard model

A new estimate is presented of the dileptonic $B$ decays $B\toπ\ell^+\ell^-(\ell=e,μ,τ)$ in naive factorization within the standard-model (SM) framework. Using a combination of several approaches, we investigate the behavior of the $B\to π$ form factors in the entire region of the momentum transfer squared $q^2$. For the vector and scalar form factors, we employ the light cone sum rule (LCSR) with a chiral current correlator to estimate, at twist-2 next-to-leading order (NLO) accuracy, their shapes in small and intermediate kinematical region. Then a simultaneous fit to a Bourrely-Caprini-Lellouch (BCL) parametrization is performed of the sum rule predictions and the corresponding lattice QCD (LQCD) results available at some high $q^2$'s. The same approach is applied for the tensor form factor, except that at large $q^2$ we use as input the LQCD data on the corresponding $B\to K$ form factor in combination with a $SU_F(3)$ symmetry breaking ansatz. Employing the fitted BCL parameterizations, we evaluate, as an illustrative example, several of the observables of the charged decay modes $B^-\toπ^-\ell^+\ell^-$, including the dilepton invariant mass distribution and branching ratio. For the dielectron and dimuon modes, the branching ratios are estimated at $\mathcal{B}(B^-\toπ^-e^+e^-)=(2.263^{+0.227}_{-0.192})\times10^{-8}$ and $\mathcal{B}(B^-\toπ^-μ^+μ^-)=(2.259^{+0.226}_{-0.191}) \times10^{-8}$. The latter shows an excellent agreement with the recent experimental measurement at LHCb and hence puts a stringent constraint on the contribution from possible new physics. We arrive at, for the ditau mode, $\mathcal{B}(B^-\toπ^-τ^+τ^-)=(1.017^{+0.118}_{-0.139})\times10^{-8}$, which is one order of magnitude larger than the existing theoretical predictions.

preprint2014arXiv

Study Majorana Neutrino Contribution to B-meson Semi-leptonic Rare Decays

B meson semi-leptonic rare decays are sensitive to new physics beyond standard model. We study the $B^{-}\to π^{-}μ^{+}μ^{-}$ process and investigate the Majorana neutrino contribution to its decay width. The constraints on the Majorana neutrino mass and mixing parameter are obtained from this decay channel with the latest LHCb data. Utilizing the best fit for the parameters, we study the lepton number violating decay $B^{-}\to π^{+}μ^{-}μ^{-}$, and find its branching ratio is about $6.4\times10^{-10}$, which is consistent with the LHCb data reported recently.

preprint2012arXiv

Form factors $f^{B\to π}_+(0)$ and $f^{D\to π}_+(0)$ in $QCD$ and determination of $|V_{ub}|$ and $|V_{cd}|$

We present a QCD study on $B, D\toπ$ semileptonic transitions at zero momentum transfer and an estimate of magnitudes of the associated CKM matrix elements. Light cone sum rules (LCSRs) with chiral correlator are applied to calculate the form factors $f^{B\to π}_+(0)$ and $f^{D\to π}_+(0)$. We show that there is no twist-3 and-5 component involved in the light-cone expansions such that the resulting sum rules have a good convergence and offer an understanding of these form factors at twist-5 level. A detailed $\mathcal{O}(α_s)$ computation is carried out in leading twist-2 approximation and the $\bar{MS}$ masses are employed for the underlying heavy quarks. With the updated inputs and experimental data, we have $f^{B\to π}_+(0)=0.28^{+0.05}_{-0.02}$ and $|V_{ub}|=(3.4^{+0.2}_{-0.6}\pm0.1\pm0.1)\times10^{-3}$; $f^{D\to π}_+(0)=0.62\pm0.03$ and $|V_{cd}|=0.244\pm0.005\pm0.003\pm0.008$. As a by-product, a numerical estimate for the decay constant $f_D$ is yielded as $f_{D}=190^{+12}_{-11}\mathrm{MeV}$.