The proceedings contain 21 papers. The topics discussed include: long waves simulation in coastal systems using parallel computational technologies;improvement of numerical solution smoothness for the hydrodynamics pr...
The proceedings contain 21 papers. The topics discussed include: long waves simulation in coastal systems using parallel computational technologies;improvement of numerical solution smoothness for the hydrodynamics problems modeling on rectangular grids;parallel implementation of substance transport problems for restoration the salinity ï¬üeld based on schemes of high order of accuracy;recognition of faces, head positions, gender, age, and emotions in realtime using deep convolutional neural networks;the quantum version of classiï¬ücation decision tree constructing algorithm C5.0;and influence of dropout and dynamic receptive field operations on convolutional networks
This paper addresses the problem of continuously finding highly correlated pairs of time series over the most recent time window. The solution builds upon the ParCorr parallel method for online correlation discovery a...
详细信息
ISBN:
(纸本)9789897583773
This paper addresses the problem of continuously finding highly correlated pairs of time series over the most recent time window. The solution builds upon the ParCorr parallel method for online correlation discovery and is designed to run continuously on top of the UPM-CEP data streaming engine through efficient streaming operators. The implementation takes advantage of the flexible API of the streaming engine that provides low level primitives for developing custom operators. Thus, each operator is implemented to process incoming tuples on-the-fly and hence emit resulting tuples as early as possible. This guarantees a real pipelined flow of data that allows for outputting early results, as the experimental evaluation shows.
Work-stealing is a key component of many parallel task graph libraries such as Intel Threading Building Blocks (TBB) FlowGraph, Microsoft Task parallel Library (TPL) Batch .Net, Cpp-Taskflow, and Nabbit. However, desi...
详细信息
ISBN:
(数字)9781728190747
ISBN:
(纸本)9781728183824
Work-stealing is a key component of many parallel task graph libraries such as Intel Threading Building Blocks (TBB) FlowGraph, Microsoft Task parallel Library (TPL) Batch .Net, Cpp-Taskflow, and Nabbit. However, designing a correct and effective work-stealing scheduler is a notoriously difficult job, due to subtle implementation details of concurrency controls and decentralized coordination between threads. This problem becomes even more challenging when striving for optimal thread usage in handling parallel workloads with complex task graphs. As a result, we introduce in this paper an effective work-stealing scheduler for execution of task dependency graphs. Our scheduler adopts a simple and efficient strategy to adapt the number of working threads to available task parallelism at any time during the graph execution. Our strategy is provably good in preventing resource underutilization and simultaneously minimizing resource waste when tasks are scarce. We have evaluated our scheduler on both micro-benchmarks and a real-world circuit timing analysis workload, and demonstrated promising results over existing methods in terms of runtime, energy efficiency, and throughput.
In this paper, we propose a distributed, unordered, label-correcting distance-1 Grundy (vertex) coloring algorithm, namely, distributed Control (DC) coloring algorithm. Our algorithm eliminates the need for vertex-cen...
详细信息
ISBN:
(纸本)9781728136134
In this paper, we propose a distributed, unordered, label-correcting distance-1 Grundy (vertex) coloring algorithm, namely, distributed Control (DC) coloring algorithm. Our algorithm eliminates the need for vertex-centric barriers and global synchronization for color refinement, relying only on atomic operations and local termination detection to update vertex color. DC proceeds optimistically, correcting the colors asynchronously as the algorithm progresses and depends on local ordering of tasks to minimize the execution of sub-optimal work. We implement our DC coloring algorithm and the well-known Jones-Plassmann algorithm and compare their performance with 4 different types of standard RMAT graphs and real-world graphs. We show that the elimination of waiting time of global and vertex-centric barriers and investing this time for local ordering leads to improved scaling for graphs with prominent power-law characteristics and densely interconnected local subgraphs.
distributed metadata management, administrating the distribution of metadata nodes on different metadata servers (MDS's), can substantially improve overall performance of large-scale distributed storage systems if...
详细信息
ISBN:
(纸本)9781450362955
distributed metadata management, administrating the distribution of metadata nodes on different metadata servers (MDS's), can substantially improve overall performance of large-scale distributed storage systems if well designed. A major difficulty confronting many metadata management schemes is the trade-off between two conflicting aspects: system load balance and metadata locality preservation. It becomes even more challenging as file access pattern inevitably varies with time. However, existing works dynamically reallocate nodes to different servers adopting history-based coarse-grained methods, failing to make timely and efficient update on distribution of nodes. In this paper, we propose an adaptive fine-grained metadata management scheme, AdaM, leveraging Deep Reinforcement Learning, to address the trade-off dilemma against time-varying access pattern. At each time step, AdaM collects environmental "states" including access pattern, the structure of namespace tree and current distribution of nodes on MDS's. Then an actor-critic network is trained to reallocate hot metadata nodes to different servers according to the observed "states". Adaptive to varying access pattern, AdaM can automatically migrate hot metadata nodes among servers to keep load balancing while maintaining metadata locality. We test AdaM on real-world data traces. Experimental results demonstrate the superiority of our proposed method over other schemes.
Due to the large scale and three-phase unbalance problem of active distribution networks (ADNs), traditional centralized state estimation (CSE) cannot meet the requirements on estimation accuracy and efficiency for re...
详细信息
Due to the large scale and three-phase unbalance problem of active distribution networks (ADNs), traditional centralized state estimation (CSE) cannot meet the requirements on estimation accuracy and efficiency for real-time analysis and control of distribution systems. This paper presents a novel consensus-based coordination distributed state estimation (CC-DSE) approach considering communication failures in ADNs. A network partition approach coupled with placement of phasor measurement units (PMUs), which is based on topology analysis, is firstly executed to divide the large-scale distribution network into several sub-regions. Then, the local state estimation (SE) of each sub-region is executed in parallel by improved weighted least squares (WLS) method with revising weights adaptive to residuals without changing the initial objective function, which can mitigate the negative effect of inner-region communication failures. The data coordination is embedded in the iteration of improved WLS of each sub-region estimation by consensus algorithm, which is robust to data missing and bad data caused by inter-region communication failures. The CC-DSE has better performance in accuracy, efficiency and robustness compared with CSE and distributed state estimation (DSE), which is demonstrated by the simulation results of the unbalanced IEEE 123 bus distribution network.
Many real-world applications feature data accesses on periodic domains. Manually implementing the synchronizations and communications associated to the data dependences on each case is cumbersome and error-prone. It i...
详细信息
Many real-world applications feature data accesses on periodic domains. Manually implementing the synchronizations and communications associated to the data dependences on each case is cumbersome and error-prone. It is increasingly interesting to support these applications in high-level parallel programming languages or parallelizing compilers. In this paper, we present a technique that, for distributed-memory systems, calculates the specific communications derived from data-parallel codes with or without periodic boundary conditions on affine access expressions. It makes transparent to the programmer the management of aggregated communications for the chosen data partition. Our technique moves to runtime part of the compile-time analysis typically used to generate the communication code for affine expressions, introducing a complete new technique that also supports the periodic boundary conditions. We present an experimental study to evaluate our proposal using several study cases. Our experimental results show that our approach can automatically obtain communication codes as efficient as those found in MPI reference codes, reducing the development effort.
Compression of floating-point data, both lossy and lossless, is a topic of increasing interest in scientific computing. Developing and evaluating suitable compression algorithms requires representative samples of data...
详细信息
Data Stream Management systems (DSMSs) performing online analytics rely on the efficient execution of large numbers of Aggregate Continuous Queries (ACQs). In this paper, we study the problem of generating high qualit...
详细信息
The proceedings contain 16 papers. The topics discussed include: ZerOBNL: a framework for distributed and reproducible co-simulation;towards an assisted simulation planning for co-simulation of cyber-physical energy s...
ISBN:
(纸本)9781728106373
The proceedings contain 16 papers. The topics discussed include: ZerOBNL: a framework for distributed and reproducible co-simulation;towards an assisted simulation planning for co-simulation of cyber-physical energy systems;cross-platform comparison of standard power system components used in realtime simulation;over current relay modeling using modelica with cross-verification against a validated model;ExSol: collaboratively assessing cybersecurity risks for protecting energy delivery systems;a digital twin for cyber-physical energy systems;automated parameter identification and calibration for the Itaipu power generation system using modelica, FMI, and RaPId;and collaborative simulation of heterogeneous components as a means toward a more comprehensive analysis of smart grids.
暂无评论