The Euclidean distance transform (EDT) is used in various methods in pattern recognition, computer vision, image analysis, physics, applied mathematics and robotics. Until now, several sequential EDT algorithms have b...
详细信息
The Euclidean distance transform (EDT) is used in various methods in pattern recognition, computer vision, image analysis, physics, applied mathematics and robotics. Until now, several sequential EDT algorithms have been described in the literature, however they are time- and memory-consuming for images with large resolutions. Therefore, parallel implementations of the EDT are required specially for 3D images. This paper presents a parallel implementation based on domain decomposition of a well-known 3D Euclidean distance transform algorithm, and analyzes its performance on a cluster of workstations. The use of a data compression tool to reduce communication time is investigated and discussed. Among the obtained performance results, this work shows that data compression is an essential tool for clusters with low-bandwidth networks.
The research on complex Brain networks plays a vital role in understanding the connectivity patterns of the human brain and disease-related alterations. Recent studies have suggested a noninvasive way to model and ana...
详细信息
We describe a distributed reasoning system called Otto-Mate that is used to detect, reason about, and respond to incidents on a computing network. Events for monitoring computingnetworks occur at different system lev...
详细信息
ISBN:
(纸本)9781424481804
We describe a distributed reasoning system called Otto-Mate that is used to detect, reason about, and respond to incidents on a computing network. Events for monitoring computingnetworks occur at different system levels. Some information might relate to data, some might be operating system specific, some application or service related, some could be network related, and from each there will be compound events that describe incident effects and information about the situation context. All together there can be thousands of events per second. Today's approaches to monitoring networks are typically centralized, sending events over the network to a single engine for analysis. Centralized monitoring ultimately cannot scale to address the volume of events that one would ideally like to be able to monitor, so techniques of today often make severe compromises relating to the events that they ingest. Centralized monitoring creates a single point of failure and also generates significant network load. To overcome these deficiencies we have developed a more distributed, approach: our reasoner agents can (in theory) be installed on every monitored resources and the reasoner language (used for programming the reasoners) enables knowledge in a reasoner's working memory to be synchronized over multiple reasoners enabling them to implement paralleldistributed reasoning algorithms that are able detect event patterns irrespective of whether the events are local or remote. Distributing the reasoning makes the system extremely resilient. Additionally, since the knowledge shared between the reasoning agents represents summary information, and because many on-line event correlation algorithms often suppress reporting once an incident has been reported, the amount of network load needed to support the distributed monitoring can actually be reduced. To demonstrate our approach we describe its application to the monitoring of a computing network that has been instrumented to protect it ag
Context-aware computing is one of the key issues in virtual computing environment. This paper aims to compare the ontology-based context modeling and reasoning approach with the traditional non-semantic one. A general...
详细信息
The proceedings contain 52 papers. The topics discussed include: improved bitonic sorting by wire elimination;parallel join processing on graphics processors for the resource description framework;modeling data distri...
ISBN:
(纸本)9783800732227
The proceedings contain 52 papers. The topics discussed include: improved bitonic sorting by wire elimination;parallel join processing on graphics processors for the resource description framework;modeling data distribution for two-phase flow problems by weighted graphs;distributed vision graph update in mobile vision networks;routing based on evolved agents;a block device driver for parallel and fault-tolerant storage;operating system processor scheduler design for future chip multiprocessor;evaluation and refinement of a tuning tool for grid applications;platform-independent modeling of explicitly parallel programs;and delivering guidance information in heterogeneous systems.
The vast quantity of data contributed and consumed via the Internet provides an environment where Collective Intelligence (CI) can emerge. This article considers CI in relation to the Peer-to-Peer (P2P) paradigm, an e...
详细信息
Idle desktops have been successfully used to run sequential and master-slave task parallel codes on a large scale in the context of volunteer computing. However, execution of message passing parallel programs in such ...
详细信息
Data aggregation is an essential yet time-consuming task in wireless sensor networks (WSNs). This paper studies the well-known Minimum-Latency Aggregation Schedule (MLAS) problem and proposes an energy-efficient distr...
详细信息
ISBN:
(纸本)9780769540597
Data aggregation is an essential yet time-consuming task in wireless sensor networks (WSNs). This paper studies the well-known Minimum-Latency Aggregation Schedule (MLAS) problem and proposes an energy-efficient distributed scheduling algorithm named Clu-DDAS based on a novel cluster-based aggregation tree. Our approach differs from all the previous schemes where Connected Dominating Sets or Maximal Independent Sets are employed. We prove that Clu-DDAS has a latency bound of 4R' + 2 Delta - 2, where. is the maximum degree and R' is the inferior network radius which is smaller than the network radius R. Clu-DDAS has comparable latency as the previously best centralized algorithm E-PAS, while Clu-DDAS consumes 78% less energy as shown by the simulation results. Clu-DDAS outperforms the previously best distributed algorithm DAS whose latency bound is 16R' + Delta - 14 on both latency and energy consumption. On average, Clu-DDAS transmits 67% fewer total messages than DAS does. We also propose an adaptive strategy for updating the schedule to accommodate dynamic network topology.
Service availability and QoS, in terms of customer affecting performance metrics, is crucial for service systems. However, the increasing complexity in distributed service systems introduce hidden space for software f...
详细信息
Cloud computing provides a framework for supporting end users easily attaching powerful services and applications through Internet. To provide secure and reliable services in cloud computing environment is an importan...
详细信息
暂无评论