The large scale of linear systems of equations results in costly solving time. These systems usually have specific properties that can be used for designing fost algorithms. In addition, using parallel programming on ...
详细信息
ISBN:
(纸本)9781424416936
The large scale of linear systems of equations results in costly solving time. These systems usually have specific properties that can be used for designing fost algorithms. In addition, using parallel programming on distributed memory clusters enables us to get the results even faster. This work introduces a new fast parallel algorithm for solving systems with a strictly diagonally dominant three-band Toeplitz coefficient matrix. We call this new method the sewing method because the boundaries sew the adjacent subsystems together.
Traditional software distributed shared memory (SDSM) systems modify the semantics of a real hardware shared memory system by relaxing the coherence semantic and by limiting the memory regions that are actually shared...
详细信息
Traditional software distributed shared memory (SDSM) systems modify the semantics of a real hardware shared memory system by relaxing the coherence semantic and by limiting the memory regions that are actually shared. These semantic modifications are done to improve performance of the applications using it. In this paper. we will show that a SDSM system that behaves like a real shared memory system (without the afore-mentioned relaxations) can also be used to execute OpenMP applications and achieve similar speedups as the ones obtained by traditional SDSM systems. This performance can be achieved by encouraging the cooperation between the SDSM and the OpenMP runtime instead of relaxing the semantics of the shared memory. In addition, techniques like boundaries alignment and page presend are demonstrated as very useful to overcome the limitations of the current SDSM systems. (c) 2005 Elsevier Inc. All rights reserved.
An examination is made of heuristic algorithms for processing distributed queries using generalized joins. As this optimization problem is NP-hard, a heuristic algorithm is used to form/formulate strategies to process...
详细信息
ISBN:
(纸本)0818608935
An examination is made of heuristic algorithms for processing distributed queries using generalized joins. As this optimization problem is NP-hard, a heuristic algorithm is used to form/formulate strategies to process queries. It has a special property in that its overhead can be controlled. The higher its overhead the better the strategies it produces. Modeling on a testbed of queries demonstrates that there is a tradeoff between the strategy's execution and formulation delays. The modeling results also support the notion that simple greedy heuristic algorithms are sufficient in that they are likely to lead to near-optimal strategies and that increasing the overhead in forming strategies is only marginally beneficial. Both the strategy formulation and execution delays are examined in relation to the number of operations specified by the strategy and the total size of partial results.
A set of mutually distrusting participants that want to agree on a common opinion must solve an instance of a Byzantine agreement problem. These problems have been extensively studied in the literature. However, most ...
详细信息
ISBN:
(纸本)9781665440660
A set of mutually distrusting participants that want to agree on a common opinion must solve an instance of a Byzantine agreement problem. These problems have been extensively studied in the literature. However, most of the existing solutions assume that the participants are aware of n - the total number of participants in the system - and f - an upper bound on the number of Byzantine participants. In this paper, we show that most of the fundamental agreement problems can be solved without affecting resiliency even if the participants do not know the values of (possibly changing) n and f. Specifically, we consider a synchronous system where the participants have unique but not necessarily consecutive identifiers, and give Byzantine agreement algorithms for reliable broadcast, approximate agreement, rotor-coordinator, early terminating consensus and total ordering in static and dynamic systems, all with the optimal resiliency of n > 3f. Moreover, we show that some synchrony is necessary as an agreement with probabilistic termination is impossible in a semi-synchronous or asynchronous system if the participants are unaware of n and f.
Modern database systems-and in particular multidatabase systems, federated database systems and OLAP-oriented data-warehouses-factor the stored data into actual content and meta-data, which structures the available in...
详细信息
ISBN:
(纸本)0769512305
Modern database systems-and in particular multidatabase systems, federated database systems and OLAP-oriented data-warehouses-factor the stored data into actual content and meta-data, which structures the available information. This separation is particularly popular in the data-warehousing community, where large amounts of data are structured by and accessible through far smaller amounts of meta-data. Given a parallel or distributed database system, a sophisticated query optimizer requires information on the specific structure of the computing resource, the data distribution, data replication and interconnection bandwidths in order to optimally schedule computational subtasks. Such optimizations are of an even greater importance for next-generation OLAP engines, which attempt to substitute on-demand aggregation into virtual data-cubes and caching for eager preaggregation. Given the demand for ever faster response times and exponentially growing data sets, the parallelization of analytical database systems continuously gains in importance. As a result a portable and comprehensive framework for the representation of optimization-related information becomes necessary. This presentation discusses the possibilities for query optimization in parallel and distributed OLAP systems, given a detailed description of the underlying computing and storage infrastructure. A modeling framework for the description of the computing resource is introduced, which may be applied to a wide array of database systems. A discussion of using profiling information gathered during the execution of queries to dynamically refine the cost estimates given in the meta-data is also provided.
This paper proposes an improved parallel control strategy which suits for the auxiliary inverter under different working conditions in metro. Firstly, differential droop coefficient and virtual impedance are introduce...
详细信息
ISBN:
(纸本)9781728124551
This paper proposes an improved parallel control strategy which suits for the auxiliary inverter under different working conditions in metro. Firstly, differential droop coefficient and virtual impedance are introduced to improve the stability and dynamic response of the system. Meanwhile, the proportional-resonant control is introduced to enhance the ability of the auxiliary inverter under the sudden load change condition or with the non-linear load. Furthermore, mathematical modeling of the auxiliary inverters is carried out in the system. Finally, the improved droop control, virtual impedance and current resonant control strategies are applied to two paralleled inverters test platform of 120kVA. The experimental results show that the proposed control algorithm can greatly improve the voltage waveform and dynamic response of the parallel inverters, and improve the ability of the system under non-linear loads.
The multi-parallel chopper system is commonly used in modern MW-level wind turbine converters. Even though the chopper is usually considered as robust and reliable, the semiconductor module can suffer critical junctio...
详细信息
ISBN:
(纸本)9781728169903
The multi-parallel chopper system is commonly used in modern MW-level wind turbine converters. Even though the chopper is usually considered as robust and reliable, the semiconductor module can suffer critical junction temperature in extreme fault ride through (FRT) events. This paper proposes a method to monitor and identify if the individual chopper is functioning, degraded, or failed by means of collecting and comparing the parallel semiconductors' temperatures during FRT events. A simulation model is created in PLCES, the accuracy of which is further validated by a dedicated experimental setup including an infrared camera. The proposed method is verified by PLECS simulations.
In this paper, the cascaded H-bridge STATCOM parallel system is taken as the research object to analyze the instability of STATCOM system caused by the interaction between parallel STATCOMs in wind farm afflux station...
详细信息
ISBN:
(纸本)9781728124551
In this paper, the cascaded H-bridge STATCOM parallel system is taken as the research object to analyze the instability of STATCOM system caused by the interaction between parallel STATCOMs in wind farm afflux station. Firstly, the inherent relation of harmonic circulation and low-frequency disturbance will be revealed by CPS-PWM theory. Then, the output impedance and transfer function of STATCOM give an insight into the interaction between STATCOMs and low-frequency resonance, and the virtual impedance method is proposed to shape the resonant peak. Finally, the correctness of theoretical analysis is verified by real-time simulation.
parallel I/O is an effective method to optimize data movement between memory and storage for many scientific applications. Poor performance of traditional disk-based file systems has led to the design of I/O libraries...
详细信息
ISBN:
(纸本)9798350337662
parallel I/O is an effective method to optimize data movement between memory and storage for many scientific applications. Poor performance of traditional disk-based file systems has led to the design of I/O libraries which take advantage of faster memory layers, such as on-node memory, present in high-performance computing (HPC) systems. By allowing caching and prefetching of data for applications alternating computation and I/O phases, a faster memory layer also provides opportunities for hiding the latency of I/O phases by overlapping them with computation phases, a technique called asynchronous I/O. Since asynchronous parallel I/O in HPC systems is still in the initial stages of development, there hasn't been a systematic study of the factors affecting its performance. In this paper, we perform a systematic study of various factors affecting the performance and efficacy of asynchronous I/O, we develop a performance model to estimate the aggregate I/O bandwidth achievable by iterative applications using synchronous and asynchronous I/O based on past observations, and we evaluate the performance of the recently developed asynchronous I/O feature of a parallel I/O library (HDF5) using benchmarks and real-world science applications. Our study covers parallel file systems on two large-scale HPC systems: Summit and Cori, the former with a GPFS storage and the latter with a Lustre parallel file system.
We study a basic information ranking problem in networks where each node holds an individual preference over a set of items and the goal for each node is to identify a sorted list of items with the largest aggregate p...
详细信息
ISBN:
(纸本)9781467325790
We study a basic information ranking problem in networks where each node holds an individual preference over a set of items and the goal for each node is to identify a sorted list of items with the largest aggregate preference. We would like to achieve this with a fully decentralized algorithm that uses a limited per-node memory and limited pair-wise communications. We show how this problem can be reduced to a plurality selection problem where the goal for each node is to identify an item with the largest aggregate ranking score, and show that solving the reduced problem solves the original ranking problem with high probability. Then we introduce a simple and natural plurality selection algorithm for the selection over m > 1 items that uses only log(2) (m) + 1 bits of per-node memory and per pair-wise communication. We prove correctness of the algorithm with high probability as the number of nodes grows large for the case when each node communicates with any other node, and establish tight convergence time bounds. The information ranking problem studied in this paper is a basic ranking problem that arises in various applications such as sorting elements in distributed computing systems, paralleldatabases, and may as well serve as a model of decentralized inference and opinion formation in distributed environments.
暂无评论