High quality video streaming for mobile users is difficult to achieve in some areas of the world due to poor broadband capacity and sparse network coverage. We propose a bandwidth-sharing scheme to allow users with li...
详细信息
Event-Based systems (EBS) can efficiently analyze large streams of sensor data in near-realtime. But they struggle with noise or incompleteness that is seen in the unprecedented amount of data generated by the Interne...
详细信息
ISBN:
(纸本)9781450332866
Event-Based systems (EBS) can efficiently analyze large streams of sensor data in near-realtime. But they struggle with noise or incompleteness that is seen in the unprecedented amount of data generated by the Internet of Things. We present a generic approach that deals with uncertain data in the middleware layer of distributed event-based systems and is hence transparent for developers. Our approach calculates alternative paths to improve the overall result of the data analysis. It dynamically generates, updates, and evaluates Bayesian Networks based on probability measures and rules defined by developers. An evaluation on position data shows that the improved detection rate justifies the computational overhead.
Large-scale graph-structured computation usually exhibits iterative and convergence-oriented computing nature, where input data is computed iteratively until a convergence condition is reached. Such features have led ...
详细信息
ISBN:
(纸本)9781450332057
Large-scale graph-structured computation usually exhibits iterative and convergence-oriented computing nature, where input data is computed iteratively until a convergence condition is reached. Such features have led to the development of two different computation modes for graph-structured programs, namely synchronous (Sync) and asynchronous (Async) modes. Unfortunately, there is currently no in-depth study on their execution properties and thus programmers have to manually choose a mode, either requiring a deep understanding of underlying graph engines, or suffering from suboptimal performance. This paper makes the first comprehensive characterization on the performance of the two modes on a set of typical graph-parallel applications. Our study shows that the performance of the two modes varies significantly with different graph algorithms, partitioning methods, execution stages, input graphs and cluster scales, and no single mode consistently outperforms the other. To this end, this paper proposes Hsync, a hybrid graph computation mode that adaptively switches a graph-parallel program between the two modes for optimal performance. Hsync constantly collects execution statistics on-the-fly and leverages a set of heuristics to predict future performance and determine when a mode switch could be profitable. We have built online sampling and offline profiling approaches combined with a set of heuristics to accurately predicting future performance in the two modes. A prototype called PowerSwitch has been built based on PowerGraph, a state-of-the-art distributed graph-parallel system, to support adaptive execution of graph algorithms. On a 48-node EC2-like cluster, PowerSwitch consistently outperforms the best of both modes, with a speedup ranging from 9% to 73% due to timely switch between two modes. Copyright 2015 ACM.
The well-orchestrated use of distilled experience, domain-specific knowledge, and well-informed trade-off decisions is imperative if we are to design effective architectures for complex software-intensive systems. In ...
The well-orchestrated use of distilled experience, domain-specific knowledge, and well-informed trade-off decisions is imperative if we are to design effective architectures for complex software-intensive systems. In particular, designing modern self-adaptive systems requires intricate decision-making over a remarkably complex problem space and a vast array of solution mechanisms. Nowadays, a large number of approaches tackle the issue of endowing software systems with self-adaptive behavior from different perspectives and under diverse assumptions, making it harder for architects to make judicious decisions about design alternatives and quality attributes trade-offs. It has currently been claimed that search-based software design approaches may improve the quality of resulting artifacts and the productivity of design processes, as a consequence of promoting a more comprehensive and systematic representation of design knowledge and preventing design bias and false intuition. To the best of our knowledge, no empirical studies have been performed to provide sound evidence of such claim in the self-adaptive systems domain. This paper reports the results of a quasi-experiment performed with 24 students of a graduate program in distributed and Ubiquitous Computing. The experiment evaluated the design of self-adaptive systems using a search-based approach proposed by us, in contrast to the use of a non-automated approach based on architectural styles catalogs. The goal was to investigate to which extent the adoption of search-based design approaches impacts on the effectiveness and complexity of resulting architectures. In addition, we also analyzed the approach’s potential for leveraging the acquisition of distilled design knowledge. Our findings show that search-based approaches can improve the effectiveness of resulting self-adaptive systems architectures and reduce their design complexity. We found no evidence regarding the approach’s potential for leveraging the acqu
The article presents the design of an application interface for associated medical data visualization and management for neurologists in a stroke clustering and prediction system called Stroke MD. The goal of the syst...
详细信息
ISBN:
(纸本)9781467375467
The article presents the design of an application interface for associated medical data visualization and management for neurologists in a stroke clustering and prediction system called Stroke MD. The goal of the system is to facilitate efficient visual data introduction and knowledge extraction based on a predictive model implementation. Aspects such as quality of the visualized data, type of the viewed data (risk factor groups, clustered groups, data alerts, conclusions etc.) and efficient user-software interaction directly influenced decisions about system design, in particular the design of the doctor's/patient's user interface sides, input methods and user-software interaction techniques, ending in to a hybrid setup of interface.
Sparse grids have become a versatile tool for a vast range of applications reaching from interpolation and numerical quadrature to data-driven problems and uncertainty *** review four selected real-world applications ...
详细信息
Sparse grids have become a versatile tool for a vast range of applications reaching from interpolation and numerical quadrature to data-driven problems and uncertainty *** review four selected real-world applications of sparse grids:financial product pricing with the Black-Scholes model,interactive explo-ration of simulation data with sparse-grid-based surrogate models,analysis of simu-lation data through sparse grid data mining methods,and stability investigations of plasma turbulence simulations.
Energy consumption has become a significant issue for data centres. Assessing their consumption requires precise and detailed models. In the latter years, many models have been proposed, but most of them either do not...
详细信息
Energy consumption has become a significant issue for data centres. Assessing their consumption requires precise and detailed models. In the latter years, many models have been proposed, but most of them either do not consider energy consumption related to virtual machine migration or do not consider the variation of the workload on (1) the virtual machines (VM) and (2) the physical machines hosting the VMs. In this paper, we show that omitting migration and workload variation from the models could lead to misleading consumption estimates. Then, we propose a new model for data centre energy consumption that takes into account the previously omitted model parameters and provides accurate energy consumption predictions for paravirtualised virtual machines running on homogeneous hosts. The new model's accuracy is evaluated with a comprehensive set of operational scenarios. With the use of these scenarios we present a comparative analysis of our model with similar state-of-the-art models for energy consumption of VM Migration, showing an improvement up to 24% in accuracy of prediction.
The increasing demand for performance has stimulated the wide adoption of many-core accelerators like Intel Xeon Phi Coprocessor, which is based on Intel's Many Integrated Core architecture. While many HPC applica...
详细信息
The increasing demand for performance has stimulated the wide adoption of many-core accelerators like Intel Xeon Phi Coprocessor, which is based on Intel's Many Integrated Core architecture. While many HPC applications running in native mode have been tuned to run efficiently on Xeon Phi, it is still unclear how a managed runtime like JVM performs on such an architecture. In this paper, we present the first measurement study of a set of Java HPC applications on Xeon Phi under JVM. One key obstacle to the study is that there is currently little support of Java for Xeon Phi. This paper presents the result based on the first porting of Open JDK platform to Xeon Phi, in which Hot Spot VM acts as the kernel execution engine. The main difficulty includes the incompatibility between Xeon Phi ISA and the assembly library of Hotspot VM. By evaluating the multithreaded Java Grande benchmark suite, we quantitatively study the performance and scalability issues of JVM on Xeon Phi and draw several conclusions from the study. To fully utilize the vector computing capability, we also present a semi-automatic vectorization in Hot Spot VM. Together with this optimization and tuning, our optimized JVM achieves up to 3.4X speedup with 60 physical cores compared to that on Xeon CPU processor. Our study indicates that it is viable and potentially performance-beneficial to run applications written for managed runtime on Xeon Phi.
This paper presents a novel approach intended to detect sinkholes in MANETs running AODV. The study focuses on the detection of thewell-known sinkhole attack, devoted to attractmost of the surrounding network traffic ...
详细信息
暂无评论