A visible nearest neighbor (VNN) query returns the k nearest objects that are visible to a query point, which is used to support various applications such as route planning, target monitoring, and antenna placement. H...
详细信息
ISBN:
(纸本)9781538637906
A visible nearest neighbor (VNN) query returns the k nearest objects that are visible to a query point, which is used to support various applications such as route planning, target monitoring, and antenna placement. However, with the proliferation of wireless communications and advances in positioning technology for mobile equipments, efficiently searching for VNN among moving objects are required. While most previous work on VNN query focused on static objects, in this paper, we treats the objects as moving consecutively when indexing them, and study the visible nearest neighbor query for moving objects (MVNN). Assuming that the objects are represented as trajectories given by linear functions of time, we propose a scheme which indexes the moving objects by time-parameterized R-tree (TPR-tree) and obstacles by R-tree. The paper offers four heuristics for visibility and space pruning. New algorithms, Post-pruning and United-pruning, are developed for efficiently solving MVNN queries with all four heuristics. The effectiveness and efficiency of our solutions are verified by extensive experiments over synthetic datasets on real road network.
The extensive growth of smartphones has spawned the propagation of malicious applications. Due to the increasing use of polymorphic malware, detection is becoming more difficult. To this end, ensemble learning has bee...
详细信息
ISBN:
(纸本)9781538637906
The extensive growth of smartphones has spawned the propagation of malicious applications. Due to the increasing use of polymorphic malware, detection is becoming more difficult. To this end, ensemble learning has been proposed to improve accuracy in malware detection, without severely sacrificing time complexity. In this paper, we propose a hybrid detection system, TFBOOST, which incorporates the tensor filter algorithm into boosting ensemble generalization architecture, in order to improve detection efficacy. TFBOOST uses a static analysis to extract features and a level-by-level boosting structure with re-sampling process to diversify base learners. Experimental results show that TFBOOST generally outperforms state-of-the-art ensemble algorithms with higher detection precision and lower false positive rates. Finally, we visually interpret the high-level results of TFBOOST and conjecture that repackaged malware is the mainstay of potential malware.
The concept of memory disaggregation has recently been gaining traction in research. With memory disaggregation, data center compute nodes can directly access memory on adjacent nodes and are therefore able to overcom...
详细信息
ISBN:
(纸本)9781665497473
The concept of memory disaggregation has recently been gaining traction in research. With memory disaggregation, data center compute nodes can directly access memory on adjacent nodes and are therefore able to overcome local memory restrictions, introducing a new data management paradigm for distributed computing. This paper proposes and demonstrates a memory disaggregated in-memory object store framework for big data applications by leveraging the newly introduced ThymesisFlow memory disaggregation system. The framework extends the functionality of the pre-existing Apache Arrow Plasma object store framework to distributed systems by enabling clients to easily and efficiently produce and consume data objects across multiple compute nodes. This allows big data applications to increasingly leverage parallelprocessing at reduced development costs. In addition, the paper includes latency and throughput measurements that indicate only a modest performance penalty is incurred for remote disaggregated memory access as opposed to local (similar to 6.5 vs similar to 5.75 GiB/s). The results can be used to guide the design of future systems that leverage memory disaggregation as well as the newly presented framework. This work is open-source and publicly accessible at https://***/10.5281/zenodo.6368998.
Many of todays important applications of our everyday lives, e.g. weather forecast, design of plane and car shapes, medical analysis or even search engine queries depend on massively-parallel computer programs that ar...
详细信息
ISBN:
(纸本)9780769561493
Many of todays important applications of our everyday lives, e.g. weather forecast, design of plane and car shapes, medical analysis or even search engine queries depend on massively-parallel computer programs that are executed in data centers hosting thousands of computers. A large amount of electrical energy is used to power them, and it is of primary importance to compute more efficiently to sustain the increasing demand of computing power while keeping energy consumption reasonable. One promising research path in this domain is heterogeneous systems. The rationale for that is that at least parts of applications execute more efficiently depending on the computing resource (processors, accelerators, etc.). Nevertheless, the exploitation of these heterogeneous platforms raises new challenges in terms of application management optimization on available computing resources. The aim of our work is to determine effective algorithms to exploit these heterogeneous platforms by finding the best mapping and scheduling of an application to optimize the execution time and energy consumption with respect to various constraints. To achieve this goal, there is a need of a detailed modeling of the applications and the underlying hardware to be able to find realistic solutions. In this paper, we propose such as model, provide two implementations with state-of-the-art tools and preliminary mapping and scheduling numerical results.
The Costas Array Problem is a highly combinatorial problem linked to radar applications. We present in this paper its detailed modeling and solving by Adaptive Search, a constraint-based local search method. Experimen...
详细信息
ISBN:
(纸本)9780769546766
The Costas Array Problem is a highly combinatorial problem linked to radar applications. We present in this paper its detailed modeling and solving by Adaptive Search, a constraint-based local search method. Experiments have been done on both sequential and parallel hardware up to several hundreds of cores. Performance evaluation of the sequential version shows results outperforming previous implementations, while the parallel version shows nearly linear speedups up to 8,192 cores.
Enterprise architects and information system designers need to understand and manage workflows, data flows, and social interactions to design tools and systems for well-coordinated organizational operations. However, ...
详细信息
ISBN:
(纸本)9781538637906
Enterprise architects and information system designers need to understand and manage workflows, data flows, and social interactions to design tools and systems for well-coordinated organizational operations. However, the organizational-nature has drastically transformed over the recent years due to wide-scale use of new computing technologies. Disintegrated structures, large quantities of frequently-generated data, and dubious system and interaction boundaries are some of the obvious identifiers of a modern enterprise, where poorly designed coordination can lead to serious privacy risks. Old coordination modeling frameworks do not set well for the new organizational settings, and a need for alternative models and frameworks has been felt. In this paper, we propose a privacy-aware conceptual framework for understanding coordination by identifying and mapping work, data, and interaction patterns in organizational environments. These propositions intend to help practitioners in developing an updated understanding of the coordination that serves privacy needs, as well.
Virtual Machine Migration (VMM) is a key technology in data centers. Due to the uncertainty of applications in resource allocation, the imbalance of resource utilization is badly poor. In this paper, an Auto-regressio...
详细信息
ISBN:
(纸本)9781538637906
Virtual Machine Migration (VMM) is a key technology in data centers. Due to the uncertainty of applications in resource allocation, the imbalance of resource utilization is badly poor. In this paper, an Auto-regression Moving Average model is proposed to predict the resource requirement of a certain virtual machine, while the resource utilization rate of the physical machine is analyzed. Since the existing VMM scheme basically follows one by one migration, which makes the migration be unable to achieve the global optimum and save more energy, the paper introduces migrated cost matrix, and recommends a set of migrations with the best performance from a global point of view to carry out the migration, and improves the "three-step" method in the traditional migrated scheme to achieve energy efficiency. Simulation experiments show that the proposed scheme in the paper can effectively reduce the energy consumption, and can improve the quality of service in a certain degree.
In order to ensure Web application QoS and remain low cost for the provider simultaneously in Web traffic burst scenario, workload prediction and corresponding provisioning mechanism are investigated in this paper. At...
详细信息
ISBN:
(纸本)9781538637906
In order to ensure Web application QoS and remain low cost for the provider simultaneously in Web traffic burst scenario, workload prediction and corresponding provisioning mechanism are investigated in this paper. At first, a virtual resource provisioning architecture is designed based on workload prediction under cloud hosting environment. Then, combined with the existing Moving Average (MA) model and Support Vector Regression (SVR) model, we propose an integrated workload prediction model named LOF-MASVR through using Local Outlier Factor (LOF) algorithm. With the application of this prediction model, a dynamic mechanism of resource provisioning is proposed by introducing an adaptive coefficient. Experiments using real Web workload demonstrate that LOF-MASVR model can reduce Service Level Agreement (SLA) violation number in peak periods around 70%, in comparison with existing prediction models. According to resource provisioning experiments on CloudSim platform, the proposed dynamic provisioning policy can effectively reduce the risk of SLA violation, which simultaneously ensures the virtual resource utilization.
As modern supercomputing systems reach the peta-flop performance range, they grow in both size and complexity. This makes them increasingly vulnerable to failures from a variety of causes. Checkpointing is a popular t...
详细信息
ISBN:
(纸本)9781424437511
As modern supercomputing systems reach the peta-flop performance range, they grow in both size and complexity. This makes them increasingly vulnerable to failures from a variety of causes. Checkpointing is a popular technique for tolerating such failures, enabling applications to periodically save their state and restart computation after a failure. Although a many automated system-level checkpointing solutions are currently available to HPC users, manual application-level checkpointing remains more popular due to its superior performance. This paper improves performance of automated checkpointing via a compiler analysis for incremental checkpointing. This analysis, which works with both sequential and OpenMP applications, reduces checkpoint sizes by as much as 80% and enables asynchronous checkpointing.
The ever-increasing supercomputer architectural complexity emphasizes the need for high-level parallel programming paradigms. Among such paradigms, task-based programming manages to abstract away much of the architect...
详细信息
ISBN:
(纸本)9781509036820
The ever-increasing supercomputer architectural complexity emphasizes the need for high-level parallel programming paradigms. Among such paradigms, task-based programming manages to abstract away much of the architecture complexity while efficiently meeting the performance challenge, even at large scale. Dynamic run-time systems are typically used to execute task-based applications, to schedule computation resource usage and memory allocations. While computation scheduling has been well studied, the dynamic management of memory resource subscription inside such run-times has however been little explored. This paper studies the cooperation between a task-based distributed application code and a run-time system engine to control the memory subscription levels throughout the execution. We show that the task paradigm allows to control the memory footprint of the application by throttling the task submission flow rate, striking a compromise between the performance benefits of anticipative task submission and the resulting memory consumption. We illustrate the benefits of our contribution on a compressed dense linear algebra distributed application.
暂无评论