To reduce the energy consumption and build a sustainable computer infrastructure now becomes a major goal of the high performance community. A number of research projects have been carried out in the field of energy-a...
详细信息
ISBN:
(纸本)9780769545769
To reduce the energy consumption and build a sustainable computer infrastructure now becomes a major goal of the high performance community. A number of research projects have been carried out in the field of energy-aware high performance computing. this paper is devoted to categorize energy-aware computing methods for the high-end computing infrastructures, such as servers, clusters, data centers, and Grids/Clouds. Based on a taxonomy of methods and system scales, this paper reviews the current status of energy-aware HPC research and summarizes open questions and research directions of software architecture for future energy-aware HPC studies.
Gigabit network technologies have made it possible to combine workstations into a distributed, massively-parallel computer system. Middleware, such as distributed shared objects (DSO), attempts to improve programmabil...
详细信息
ISBN:
(纸本)0818678135
Gigabit network technologies have made it possible to combine workstations into a distributed, massively-parallel computer system. Middleware, such as distributed shared objects (DSO), attempts to improve programmability of such systems, by providing globally accessible 'object' abstractions, Researchers have developed consistency protocols for replicated 'memory' objects. these protocols are well suited to scientific applications but less suited to multimedia or groupware applications. this paper addresses the state sharing needs of complex distributed applications with (1) high-frequency symmetric accesses to shared objects, (2) unpredictable and limited locality of accesses, (3) dynamically changing sharing behavior, and (4) potential data races. We show that a DSO system exploiting application-level temporal and spatial constraints an shared objects can outperform shared object protocols which do not exploit application-level constraints. We compare our S(emantic) DSO against entry consistency using a sample application having the four properties mentioned above.
this paper describes the network communication occurring within a large-scale, interactive, distributed 3D environment, which supports closed coupled collaboration. Its main goals are the maintenance of low latency du...
详细信息
ISBN:
(纸本)1880843528
this paper describes the network communication occurring within a large-scale, interactive, distributed 3D environment, which supports closed coupled collaboration. Its main goals are the maintenance of low latency during user interaction and fast multicasting of messages in order to fulfill consistency requirements. this is achieved by sophisticated message distribution techniques peer-to-peer connections between interacting clients and a global hierarchical communication topology.
Band selection is a common technique for dimensionality reduction of hyperspectral imagery. When the desired object information is unknown, an unsupervised band selection approach is employed to select the most distin...
详细信息
ISBN:
(纸本)9780769545769
Band selection is a common technique for dimensionality reduction of hyperspectral imagery. When the desired object information is unknown, an unsupervised band selection approach is employed to select the most distinctive and informative bands. However, it may be time-consuming for unsupervised band selection methods that need to take all pixels into consideration. Here, we propose an approach to select several pixels for unsupervised band selection and the number of pixels required can be equal to the number of bands to be selected minus 1. With whitened pixel signatures (not the original pixels), band selection performance can be comparable to or even better than that from using all the pixels. For this approach, graphics processing unit (GPU)-based parallelcomputing is implemented for pixel selection only to further expedite the process, since computational complexity in band selection has been greatly reduced.
the growing demand for high-performance data processing stimulates the development of in-memory file systems, which exploit the advanced features of emerging non-volatile memory techniques for achieving high-speed fil...
详细信息
ISBN:
(纸本)9781509050819
the growing demand for high-performance data processing stimulates the development of in-memory file systems, which exploit the advanced features of emerging non-volatile memory techniques for achieving high-speed file accesses. Existing in-memory file systems, however, are all designed for the systems with uniformed memory accesses. their performance is poor on Non-Uniform Memory Access (NUMA) machines as they do not consider the asymmetric memory access speed and the architecture of multiple nodes. In this paper, we propose a new design of NUMA-aware in-memory file systems. We propose a distributed file system layout for leveraging the loads of in-memory file accesses on different nodes, a thread-file binding algorithm and a buffer assignment technique for increasing local memory accesses during run-time. Based on the proposed techniques, we implement a functional NUMA-aware in-memory file system, HydraFS, in Linux kernel. Extensive experiments are conducted withthe standard benchmark. the experimental results show that HydraFS significantly outperforms typical existing in-memory file systems, including EXT4-DAX, PMFS, and SIMFS.
Grid computing uses a job submission model that requires the users to perform a set of interactive operations for executing an application on the Grid. Grid users are therefore burdened withthe tasks of understanding...
详细信息
ISBN:
(纸本)9780769545769
Grid computing uses a job submission model that requires the users to perform a set of interactive operations for executing an application on the Grid. Grid users are therefore burdened withthe tasks of understanding the basic concept of Grid computing and the details of job management. Cloud computing, on the other hand, applies a utility model that allows the user to access the underlying platform via Web services. this work brings the Cloud concept to the Grid with a result of replacing the job submission model with a service infrastructure. In this case, Grid applications are presented as Web services that can be executed on the Grid automatically without user interactions. All issues related to job management are performed by the system. the service infrastructure significantly simplifies the users' task in accessing the Grid.
We investigate the problem of developing and optimizing a dynamic load. balancing scheduler for efficient execution of Logic and Constraint programs on Beowulf clusters. We experimentally evaluate alternative scheduli...
详细信息
ISBN:
(纸本)1880843528
We investigate the problem of developing and optimizing a dynamic load. balancing scheduler for efficient execution of Logic and Constraint programs on Beowulf clusters. We experimentally evaluate alternative scheduling strategies (centralized vs. distributed, top-most vs. bottommost, random vs. non-random) and use the results to implement and optimize the scheduler used in PALS-an efficient distributed implementation of Prolog.
Spatial data processing in general involves large-scale distributedcomputing resources across multiple data centers and distributed massive data sets. Task scheduling of spatial data processing across distributed dat...
详细信息
ISBN:
(纸本)9780769545769
Spatial data processing in general involves large-scale distributedcomputing resources across multiple data centers and distributed massive data sets. Task scheduling of spatial data processing across distributed data centers brings new research issues and calls for novel scheduling mechanisms and frameworks. this paper analyzes the features of spatial data processing, identifies the research issues of task scheduling in massive spatial data processing, and proposes task model & scheduling mechanisms for task scheduling in spatial data processing.
the distant data centre-centric Internet of things systems face the latency issue especially in the real-time-based applications. Recently, Fog computing models have been introduced to overcome the latency issue by ut...
详细信息
ISBN:
(纸本)9781509050819
the distant data centre-centric Internet of things systems face the latency issue especially in the real-time-based applications. Recently, Fog computing models have been introduced to overcome the latency issue by utilising the proximity based computational resources. However, the increasing users of Fog computing servers will cause bottleneck issues and consequently the latency issue arises again. this paper introduces the utilisation of Mist computing (Mist) model, which exploits the computational and networking resources from the devices at the very edge of IoT networks. the proposed service-oriented mobile-embedded Platform as a Service framework enables the edge IoT devices to provide a platform that allows requesters to deploy and execute their own program models. the framework supports resource-aware autonomous service configuration that can manage the availability of the functions provided by the Mist node based on the dynamically changing hardware resource availability. Additionally, the framework also supports task distribution among a group of Mist nodes. the prototype has been tested and performance evaluated on the real world devices.
In recent years, cloud computing has gradually become the mainstream of Internet services. When cloud computing environments become more perfect, the business and user will be an enormous amount of data stored in the ...
详细信息
ISBN:
(纸本)9780769545769
In recent years, cloud computing has gradually become the mainstream of Internet services. When cloud computing environments become more perfect, the business and user will be an enormous amount of data stored in the remote cloud storage devices, hoping to achieve random access, data collection, reduce costs, facilitate the sharing of other services. However, when the data is stored in the cloud storage device, a long time, enterprises and users inevitably will have security concerns, fearing that the information is actually stored in the cloud is still in the storage device or too long without access to, has long been the cloud server removed or destroyed, resulting in businesses and users in the future can't access or restore the data files. therefore, this scheme goal to research and design for data storage cloud computing environments that are proved. Stored in the cloud for data storage, research and develop a security and efficient storage of proof protocol, also can delegate or authorize others to public verifiability whether the data actually stored in the cloud storage devices.
暂无评论