Searcharxiv⌕ Search

arXiv subjects

Rui Han

Publications and source records attributed to Rui Han.

At least 91 records · Page 5Linked to original sources
arXiv2017

AccurateML: Information-aggregation-based Approximate Processing for Fast and Accurate Machine Learning on MapReduce

The growing demands of processing massive datasets have promoted irresistible trends of running machine learning applications on MapReduce. When processing large input data, it is often of greater values to produce fast and accurate enough approximate results than slow exact results. Existing techniques produce approximate results by processing parts of the input data, thus incurring large accuracy losses when using short job execution times, because all the skipped input data potentially contributes to result accuracy. We address this limitation by proposing AccurateML that aggregates information of input data in each map task to create small aggregated data points. These aggregated points enable all map tasks producing initial outputs quickly to save computation times and decrease the outputs' size to reduce communication times. Our approach further identifies the parts of input data most related to result accuracy, thus first using these parts to improve the produced outputs to minimize accuracy losses. We evaluated AccurateML using real machine learning applications and datasets. The results show: (i) it reduces execution times by 30 times with small accuracy losses compared to exact results; (ii) when using the same execution times, it achieves 2.71 times reductions in accuracy losses compared to existing approximate processing techniques.

cs.DC↗
arXiv2017

Discrete Bethe-Sommerfeld Conjecture

In this paper, we prove a discrete version of the Bethe-Sommerfeld conjecture. Namely, we show that the spectra of multi-dimensional discrete periodic Schrödinger operators on $\mathbb{Z}^d$ lattice with sufficiently small potentials contain at most two intervals. Moreover, the spectrum is a single interval, provided one of the periods is odd, and can have a gap whenever all periods are even.

math-ph↗
arXiv2016

Uniform localization is always uniform

In this note we show that if a family of ergodic Schrödinger operators on $l^2({\Bbb Z}^γ)$ with continuous potentials have uniformly localized eigenfunctions then these eigenfunctions must be uniformly localized in a homogeneous sense.

math.SP↗
arXiv2016

Quantum dynamical bounds for ergodic potentials with underlying dynamics of zero topological entropy

In this paper we obtain upper quantum dynamical bounds as a corollary of positive Lyapunov exponent for Schrödinger operators $H_{f,θ} u(n)=u(n+1)+u(n-1)+ ϕ(f^nθ)u(n)$, where $ϕ: \mathcal{M}\to {\Bbb R}$ is a piecewise Hölder function on a compact Riemannian manifold $\mathcal{M}$, and $f:\mathcal{M}\to\mathcal{M}$ is a uniquely ergodic volume preserving map with zero topological entropy. As corollaries we obtain localization-type statements for shifts and skew-shifts on higher dimensional tori with arithmetic conditions on the parameters. These are the first localization-type results with precise arithmetic conditions for multi-frequency quasiperiodic and skew-shift potentials.

math.SP↗
arXiv2016

AccuracyTrader: Accuracy-aware Approximate Processing for Low Tail Latency and High Result Accuracy in Cloud Online Services

Modern latency-critical online services such as search engines often process requests by consulting large input data spanning massive parallel components. Hence the tail latency of these components determines the service latency. To trade off result accuracy for tail latency reduction, existing techniques use the components responding before a specified deadline to produce approximate results. However, they may skip a large proportion of components when load gets heavier, thus incurring large accuracy losses. This paper presents AccuracyTrader that produces approximate results with small accuracy losses while maintaining low tail latency. AccuracyTrader aggregates information of input data on each component to create a small synopsis, thus enabling all components producing initial results quickly using their synopses. AccuracyTrader also uses synopses to identify the parts of input data most related to arbitrary requests' result accuracy, thus first using these parts to improve the produced results in order to minimize accuracy losses. We evaluated AccuracyTrader using workloads in real services. The results show: (i) AccuracyTrader reduces tail latency by over 40 times with accuracy losses of less than 7% compared to existing exact processing techniques; (ii) when using the same latency, AccuracyTrader reduces accuracy losses by over 13 times comparing to existing approximate processing techniques.

cs.DC↗
arXiv2016

Implementing a neutral-atom controlled-phase gate with a single Rydberg pulse

One can implement fast two-qubit entangling gates by exploiting the Rydberg blockade. Although various theoretical schemes have been proposed, experimenters have not yet been able to demonstrate two-atom gates of high fidelity due to experimental constraints. We propose a novel scheme, which only uses a single Rydberg pulse illuminating both atoms, for the construction of neutral-atom controlled-phase gates. In contrast to the existing schemes, our approach is simpler to implement and requires neither individual addressing of atoms nor adiabatic procedures. With parameters estimated based on actual experimental scenarios, a gate fidelity higher than 0.99 is achievable.

quant-ph↗
arXiv2015

BigDataBench-MT: A Benchmark Tool for Generating Realistic Mixed Data Center Workloads

Long-running service workloads (e.g. web search engine) and short-term data analysis workloads (e.g. Hadoop MapReduce jobs) co-locate in today's data centers. Developing realistic benchmarks to reflect such practical scenario of mixed workload is a key problem to produce trustworthy results when evaluating and comparing data center systems. This requires using actual workloads as well as guaranteeing their submissions to follow patterns hidden in real-world traces. However, existing benchmarks either generate actual workloads based on probability models, or replay real-world workload traces using basic I/O operations. To fill this gap, we propose a benchmark tool that is a first step towards generating a mix of actual service and data analysis workloads on the basis of real workload traces. Our tool includes a combiner that enables the replaying of actual workloads according to the workload traces, and a multi-tenant generator that flexibly scales the workloads up and down according to users' requirements. Based on this, our demo illustrates the workload customization and generation process using a visual interface. The proposed tool, called BigDataBench-MT, is a multi-tenant version of our comprehensive benchmark suite BigDataBench and it is publicly available from http://prof.ict.ac.cn/BigDataBench/multi-tenancyversion/.

cs.DC↗
arXiv2015

Investigations into Elasticity in Cloud Computing

The pay-as-you-go model supported by existing cloud infrastructure providers is appealing to most application service providers to deliver their applications in the cloud. Within this context, elasticity of applications has become one of the most important features in cloud computing. This elasticity enables real-time acquisition/release of compute resources to meet application performance demands. In this thesis we investigate the problem of delivering cost-effective elasticity services for cloud applications. Traditionally, the application level elasticity addresses the question of how to scale applications up and down to meet their performance requirements, but does not adequately address issues relating to minimising the costs of using the service. With this current limitation in mind, we propose a scaling approach that makes use of cost-aware criteria to detect the bottlenecks within multi-tier cloud applications, and scale these applications only at bottleneck tiers to reduce the costs incurred by consuming cloud infrastructure resources. Our approach is generic for a wide class of multi-tier applications, and we demonstrate its effectiveness by studying the behaviour of an example electronic commerce site application. Furthermore, we consider the characteristics of the algorithm for implementing the business logic of cloud applications, and investigate the elasticity at the algorithm level: when dealing with large-scale data under resource and time constraints, the algorithm's output should be elastic with respect to the resource consumed. We propose a novel framework to guide the development of elastic algorithms that adapt to the available budget while guaranteeing the quality of output result, e.g. prediction accuracy for classification tasks, improves monotonically with the used budget.

cs.DC↗
arXiv2015

PCS: Predictive Component-level Scheduling for Reducing Tail Latency in Cloud Online Services

Modern latency-critical online services often rely on composing results from a large number of server components. Hence the tail latency (e.g. the 99th percentile of response time), rather than the average, of these components determines the overall service performance. When hosted on a cloud environment, the components of a service typically co-locate with short batch jobs to increase machine utilizations, and share and contend resources such as caches and I/O bandwidths with them. The highly dynamic nature of batch jobs in terms of their workload types and input sizes causes continuously changing performance interference to individual components, hence leading to their latency variability and high tail latency. However, existing techniques either ignore such fine-grained component latency variability when managing service performance, or rely on executing redundant requests to reduce the tail latency, which adversely deteriorate the service performance when load gets heavier. In this paper, we propose PCS, a predictive and component-level scheduling framework to reduce tail latency for large-scale, parallel online services. It uses an analytical performance model to simultaneously predict the component latency and the overall service performance on different nodes. Based on the predicted performance, the scheduler identifies straggling components and conducts near-optimal component-node allocations to adapt to the changing performance interferences from batch jobs. We demonstrate that, using realistic workloads, the proposed scheduler reduces the component tail latency by an average of 67.05\% and the average overall service latency by 64.16\% compared with the state-of-the-art techniques on reducing tail latency.

cs.DC↗
arXiv2015

Characterization and Architectural Implications of Big Data Workloads

Big data areas are expanding in a fast way in terms of increasing workloads and runtime systems, and this situation imposes a serious challenge to workload characterization, which is the foundation of innovative system and architecture design. The previous major efforts on big data benchmarking either propose a comprehensive but a large amount of workloads, or only select a few workloads according to so-called popularity, which may lead to partial or even biased observations. In this paper, on the basis of a comprehensive big data benchmark suite---BigDataBench, we reduced 77 workloads to 17 representative workloads from a micro-architectural perspective. On a typical state-of-practice platform---Intel Xeon E5645, we compare the representative big data workloads with SPECINT, SPECCFP, PARSEC, CloudSuite and HPCC. After a comprehensive workload characterization, we have the following observations. First, the big data workloads are data movement dominated computing with more branch operations, taking up to 92% percentage in terms of instruction mix, which places them in a different class from Desktop (SPEC CPU2006), CMP (PARSEC), HPC (HPCC) workloads. Second, corroborating the previous work, Hadoop and Spark based big data workloads have higher front-end stalls. Comparing with the traditional workloads i. e. PARSEC, the big data workloads have larger instructions footprint. But we also note that, in addition to varied instruction-level parallelism, there are significant disparities of front-end efficiencies among different big data workloads. Third, we found complex software stacks that fail to use state-of-practise processors efficiently are one of the main factors leading to high front-end stalls. For the same workloads, the L1I cache miss rates have one order of magnitude differences among diverse implementations with different software stacks.

cs.DC↗
arXiv2015

Benchmarking Big Data Systems: State-of-the-Art and Future Directions

The great prosperity of big data systems such as Hadoop in recent years makes the benchmarking of these systems become crucial for both research and industry communities. The complexity, diversity, and rapid evolution of big data systems gives rise to various new challenges about how we design generators to produce data with the 4V properties (i.e. volume, velocity, variety and veracity), as well as implement application-specific but still comprehensive workloads. However, most of the existing big data benchmarks can be described as attempts to solve specific problems in benchmarking systems. This article investigates the state-of-the-art in benchmarking big data systems along with the future challenges to be addressed to realize a successful and efficient benchmark.

cs.PF↗
arXiv2014

Neutron Time-Of-Flight Spectrometer Based on HIRFL for Studies of Spallation Reactions Related to ADS Project

A Neutron Time-Of-Flight (NTOF) spectrometer based on Heavy Ion Research Facility in Lanzhou (HIRFL) is developed for studies of neutron production of proton induced spallation reactions related to the ADS project. After the presentation of comparisons between calculated spallation neutron production double-differential cross sections and the available experimental one, a detailed description of NTOF spectrometer is given. Test beam results show that the spectrometer works well and data analysis procedures are established. The comparisons of the test beam neutron spectra with those of GEANT4 simulations are presented.

physics.ins-det↗
arXiv2014

Characterizing and Subsetting Big Data Workloads

Big data benchmark suites must include a diversity of data and workloads to be useful in fairly evaluating big data systems and architectures. However, using truly comprehensive benchmarks poses great challenges for the architecture community. First, we need to thoroughly understand the behaviors of a variety of workloads. Second, our usual simulation-based research methods become prohibitively expensive for big data. As big data is an emerging field, more and more software stacks are being proposed to facilitate the development of big data applications, which aggravates hese challenges. In this paper, we first use Principle Component Analysis (PCA) to identify the most important characteristics from 45 metrics to characterize big data workloads from BigDataBench, a comprehensive big data benchmark suite. Second, we apply a clustering technique to the principle components obtained from the PCA to investigate the similarity among big data workloads, and we verify the importance of including different software stacks for big data benchmarking. Third, we select seven representative big data workloads by removing redundant ones and release the BigDataBench simulation version, which is publicly available from http://prof.ict.ac.cn/BigDataBench/simulatorversion/.

cs.PF↗
arXiv2014

BDGS: A Scalable Big Data Generator Suite in Big Data Benchmarking

Data generation is a key issue in big data benchmarking that aims to generate application-specific data sets to meet the 4V requirements of big data. Specifically, big data generators need to generate scalable data (Volume) of different types (Variety) under controllable generation rates (Velocity) while keeping the important characteristics of raw data (Veracity). This gives rise to various new challenges about how we design generators efficiently and successfully. To date, most existing techniques can only generate limited types of data and support specific big data systems such as Hadoop. Hence we develop a tool, called Big Data Generator Suite (BDGS), to efficiently generate scalable big data while employing data models derived from real data to preserve data veracity. The effectiveness of BDGS is demonstrated by developing six data generators covering three representative data types (structured, semi-structured and unstructured) and three data sources (text, graph, and table data).

cs.DB↗
arXiv2014

On Big Data Benchmarking

Big data systems address the challenges of capturing, storing, managing, analyzing, and visualizing big data. Within this context, developing benchmarks to evaluate and compare big data systems has become an active topic for both research and industry communities. To date, most of the state-of-the-art big data benchmarks are designed for specific types of systems. Based on our experience, however, we argue that considering the complexity, diversity, and rapid evolution of big data systems, for the sake of fairness, big data benchmarks must include diversity of data and workloads. Given this motivation, in this paper, we first propose the key requirements and challenges in developing big data benchmarks from the perspectives of generating data with 4V properties (i.e. volume, velocity, variety and veracity) of big data, as well as generating tests with comprehensive workloads for big data systems. We then present the methodology on big data benchmarking designed to address these challenges. Next, the state-of-the-art are summarized and compared, following by our vision for future research directions.

cs.PF↗
arXiv2013

Beyond adiabatic elimination: A hierarchy of approximations for multi-photon processes

In multi-level systems, the commonly used adiabatic elimination is a method for approximating the dynamics of the system by eliminating irrelevant, non-resonantly coupled levels. This procedure is, however, somewhat ambiguous and it is not clear how to improve on it systematically. We use an integro-differential equation for the probability amplitudes of the levels of interest, which is equivalent to the original Schrodinger equation for all probability amplitudes. In conjunction with a Markov approximation, the integro-differential equation is then used to generate a hierarchy of approximations, in which the zeroth order is the adiabatic-elimination approximation. It works well with a proper choice of interaction picture; the procedure suggests criteria for optimizing this choice. The first-order approximation in the hierarchy provides significant improvements over standard adiabatic elimination, without much increase in complexity, and is furthermore not so sensitive to the choice of interaction picture. We illustrate these points with several examples.

quant-ph↗
arXiv2012

Raman transitions without adiabatic elimination: A simple and accurate treatment

Driven Raman processes --- nearly resonant two-photon transitions through an intermediate state that is non-resonantly coupled and does not acquire a sizeable population --- are commonly treated with a simplified description in which the intermediate state is removed by adiabatic elimination. While the adiabatic-elimination approximation is reliable when the detuning of the intermediate state is quite large, it cannot be trusted in other situations, and it does not allow one to estimate the population in the eliminated state. We introduce an alternative method that keeps all states in the description, without increasing the complexity by much. An integro-differential equation of Lippmann-Schwinger type generates a hierarchy of approximations, but very accurate results are already obtained in the lowest order.

quant-ph↗