the use of stream processing for state monitoring of distributed infrastructures has been advocated by some in order to overcome the issues of traditional monitoring solutions when tasked with complex continuous queri...
详细信息
ISBN:
(纸本)9789897580369
the use of stream processing for state monitoring of distributed infrastructures has been advocated by some in order to overcome the issues of traditional monitoring solutions when tasked with complex continuous queries. However, in the domain of behavior monitoring the situation gets more complicated. It is mainly because of the low-quality source of behavior-related monitoring information (natural language computer logs). Existing approaches prevalently rely on indexing and real-time data-mining of the behavior-related data rather than on using event/stream processing techniques and the many corresponding benefits. the goal of this paper is to present a general notion of distributed Event-Driven Monitoring Architecture that will enable an easy definition of expressive continuous queries over many distributed and heterogeneous streams of behavior-related (and state-related) monitoring data.
In this paper a novel parallel algorithm for the tensor based classifiers for object recognition in digital images is presented. Classification is performed with an ensemble of base classifiers, each operating in the ...
详细信息
ISBN:
(纸本)9789897581335
In this paper a novel parallel algorithm for the tensor based classifiers for object recognition in digital images is presented. Classification is performed with an ensemble of base classifiers, each operating in the orthogonal subspaces obtained withthe Higher-Order Singular Value Decomposition (HOSVD) of the prototype pattern tensors. parallelism of the system is realized through the functional and data decompositions on different levels of computations. First, the parallel implementation of the HOSVD is presented. then, the second level of parallelism is gained by partitioning the input dataset. Each of the partitions is used to train a separate tensor classifiers of the ensemble. Despite the computational speed-up and lower memory requirements, also accuracy of the ensemble showed to be higher compared to a single classifier. the method was tested in the context of object recognition in computer vision. the experiments show high accuracy and accelerated performance both in the training and classification stages.
In this work, we develop an IPv6 enabled smart building test-bed facility, by combining sensing and communication devices and functionalities. We address the Internet of things paradigm by using diverse heterogeneous ...
详细信息
ISBN:
(纸本)9781479946181
In this work, we develop an IPv6 enabled smart building test-bed facility, by combining sensing and communication devices and functionalities. We address the Internet of things paradigm by using diverse heterogeneous devices such as smartphones, sensor motes, NFC technology and traditional electrical devices, each one serving a specific role in the test-bed facility. Also, we extend a basic actuation component by making it self-aware, in terms of supported resources. those enhancements allow us to enrich the test-bed's capabilities in terms of M2M communication, portability and decentralization of the actuation process. Finally, we provide a simple smart room scenario for a tunable combination of energy efficiency and comfort, which automatically adjusts the room's light level based on ambient conditions and user preferences and demonstrate the feasibility of our system.
HiPro-CodeGen is a code generation engine designed for numerical simulation development. Its central objective is to produce a parallel software framework with standard structure for an application developed on JASMIN...
详细信息
ISBN:
(纸本)9789897580369
HiPro-CodeGen is a code generation engine designed for numerical simulation development. Its central objective is to produce a parallel software framework with standard structure for an application developed on JASMIN, a domain-specific computational framework. the unique parallel part and all interfaces of the application are generated and implementation of sequential subroutines is the only part of the code left to be written manually for a programmer. the design and implementation of the code generation engine is introduced which combines numerical mathematics with component-based programming to create ontological models for parallel simulations. A hybrid programming method is proposed on the work mechanism of the engine which combines graphical and textual approaches to hide parallel programming and object-oriented programming from developers. A real application is presented to show the effectiveness and efficiency of the engine.
As China advances toward urbanization, the relation between supply and demand of land use is growing acutelyEffective supervision of various land use has become necessary to achieve reasonable and lawful use of landth...
详细信息
ISBN:
(纸本)9783319483535
As China advances toward urbanization, the relation between supply and demand of land use is growing acutelyEffective supervision of various land use has become necessary to achieve reasonable and lawful use of landthe change of land use from farmland or unused land to construction land use in the process of urbanization usually undergoes the following stages: approval,expropriation, provision, application, and supplementthe change to construction land use similarly undergoes this process, and then the outcome before and after the change is compared to determine potential land use problemsthe present process-oriented spatio-temporal data model for organization and supervision data management records the supervised spatio-temporal variation process and state by using new-added auxiliary tablesthis model can ef ficiently organize and demonstrate multiple relationships among landuse stages and the evolution process in the life cycle of spatial entitiesTo achieve effective supervision from the upper to the lower in the three-level "province-city-county" distributed environment and to detect potential land use problems early, data from different places are collected and synthesized for analysisTo ensure that the database is up-to-date,real-time synchronization of spatio-temporal supervision data should be studied and the integrity of the spatio-temporal data model should be ***, this study introduces an updating mechanism based on the trigger technology for data synchronizationWhen importing basic data from distributed databases into the supervision database, simultaneous updating in related auxiliary tables is performed along with changes in the basic tablesthat is, records are added automatically into auxiliary tables that reflect the spatio-temporal change of the process and state when basic data are importedthis method not only achieves and maintains consistency in the database, but also ensures integrity and currency in the auxiliary tablesChange in
Object detection is an important function for intelligent multimedia processing, but its computational complexity prevented its pervasive uses in consumer electronics. To process large-scale datasets in real-time, mor...
详细信息
ISBN:
(纸本)9781479967865
Object detection is an important function for intelligent multimedia processing, but its computational complexity prevented its pervasive uses in consumer electronics. To process large-scale datasets in real-time, more resources and reliable infrastructures are required for spreading the data and running the applications across multiple machines in parallel. In order to detect a large number of objects in real-time, a task-parallel processing framework based on Storm is proposed.
We present an algorithm for reconstruction of dynamic scenes from a video input of an asynchronous stereopair. Our method presumes that the timestamps of the frame acquisitions and the intrinsic parameters of the ster...
详细信息
ISBN:
(纸本)9789897581335
We present an algorithm for reconstruction of dynamic scenes from a video input of an asynchronous stereopair. Our method presumes that the timestamps of the frame acquisitions and the intrinsic parameters of the stereo cameras are known, as well as, that the extrinsic parameters can be estimated from static regions of the scene. It computes the 3D trajectory of each individual point by modeling its motion parameters. Our algorithm requires in addition to the two asynchronous images from the camera pair, two additional images from one of the cameras for its processing. Since dynamic scene elements cause wrong reconstructions, here we introduce a two stage process in which the directions of the possible motion components are estimated prior to the 3D pose estimation of the corresponding point.
Today multicore processors are ubiquitous in desktop computers, servers and various other devices. In order to take advantage of such multicore processors many of today9;s existing applications, which typically are...
详细信息
ISBN:
(纸本)9789897580369
Today multicore processors are ubiquitous in desktop computers, servers and various other devices. In order to take advantage of such multicore processors many of today's existing applications, which typically are sequential applications, need to be ported to the multicore platform. However, the adoption of multicore technology in software applications is still restrained by technical and economical obstacles. the contribution of this paper is a methodology for porting sequential software to the multicore platform. It takes into account the technical specifics of parallel programming and multicore technology offering developers orientation during the porting process. In addition to that (and in contrast to existing methodologies) it also addresses the economical obstacles of multicore adoption in software development by (1) supporting planning and cost control to counteract high development costs and by (2) utilizing auto tuning in order to cope with uncertainty due to varying processor architectures.
the proceedings contain 56 papers. the topics discussed include: nested atomic sections withthread escape: an operational semantics;RDF data query and management method based on HBase and structure index in railway s...
ISBN:
(纸本)9781479924189
the proceedings contain 56 papers. the topics discussed include: nested atomic sections withthread escape: an operational semantics;RDF data query and management method based on HBase and structure index in railway sensor application;SimpleLock: fast and accurate hybrid data race detector;improving blocking operation support in Intel TBB;implementation of a cloud IaaS with dynamic resource allocation method using OpenStack;automatic knowledge extraction supported by semantic enrichment in medical records;study on quantum parallel processing by adiabatic quantum computation in Simon problem;CADM: a centralized administration and dynamic monitoring framework for network intrusion detection based on virtualization;cHPP controller: a high performance hyper-node hardware accelerator;an ad-hoc method with genetic algorithm for printed circuit board assembly optimization on the sequential pick-and-place machine;and HRUL: a hardware assisted recorder for user-level application.
Mobile applications envisaged in the Smart City rely on the capacity of acquiring context from a vast array of sensors and other data sources, to deliver context-aware services at an urban scale whilst leveraging the ...
详细信息
ISBN:
(纸本)9781479986613
Mobile applications envisaged in the Smart City rely on the capacity of acquiring context from a vast array of sensors and other data sources, to deliver context-aware services at an urban scale whilst leveraging the Internet of things (IoT). this work analyses existing context dissemination techniques, encompassing both probabilistic and deterministic algorithms, and advocates the case for ACT (Adaptive Context Tries), a novel urban scale technique with a dissemination complexity and context availability that fulfils the requirements of Context-Aware Smart City applications. A number of enabling technologies and middleware solutions already exists supporting the provision and the sharing of context in small scale scenarios. It also shows that existing context dissemination techniques used in State of the Art middleware do not provide solutions that support context-aware applications in urban scale and dynamic, mobile environments. Results show ACT provides a scalable approach that incurs logarithmic growth in overhead with respect to the number of participating peers in the system without compromising the completeness, reliability, or timeliness of disseminated context.
暂无评论