this paper presents control implementation methods for an original distributed program design framework PEGASUS DA (Program Execution Governed by Asynchronous SUpervision of States in distributedapplications) which p...
详细信息
ISBN:
(纸本)9783642552243
this paper presents control implementation methods for an original distributed program design framework PEGASUS DA (Program Execution Governed by Asynchronous SUpervision of States in distributedapplications) which provides automated design of distributed program execution control based on program global states monitoring. the framework includes a built in support for handling local and global application states as well as automatic construction and use of strongly consistent application global states for program execution control. In particular, the paper presents methods used to implement distributed program control inside the PEGASUS DA framework run on clusters of contemporary multicore processors based on multithreading. the program design method is illustrated on a distributed multithreaded application executed with load balancing in a multicore system.
this paper proposes an intelligent framework to accurately analyze these echo images in order to discover disease category and assess the severity automatically. Typically, each video consists of 90-100 frames of 2D e...
详细信息
Several MultiConnect technologies are actively discussed in research today. MultiPath TCP (MPTCP) is capable of splitting one flow into subflows and balance the load across multiple access technologies. Multihoming is...
详细信息
ISBN:
(纸本)9783319115689
Several MultiConnect technologies are actively discussed in research today. MultiPath TCP (MPTCP) is capable of splitting one flow into subflows and balance the load across multiple access technologies. Multihoming is an older technology that makes it possible for network providers to balance load across multiple up- and down-links dynamically. Finally, Software Defined Networking (SDN) achieves the ultimate flexibility of connection and routing decisions. However, none of these technologies enable true (network or otherwise) resource-pooling in communications within arbitrary size user groups such as occur in meetings, class discussions, and ad-hoc communities in the wild. this paper proposes the concept of a Virtual Wireless User (VWU) which represents the entire group and appears as single user to an over-the-network service. Each group member is capable of MultiConnect using Wi-Fi Direct in parallel with any other connection method. Modeling based on real measurements shows that VWUs can achieve throughput in the order of tens of Mbps even if throughput of individual users is very low. the paper also formulates a formal optimization problem in relation to VWU.
Modern genotyping technologies are able to obtain up to a few million genetic markers (such as SNPs) of an individual within a few minutes of time. Detecting epistasis, such as SNP-SNP interactions, in Genome-Wide Ass...
详细信息
Multigrid methods are among the fastest numerical algorithms for solving large sparse linear systems. the Conjugate Gradient method with Multigrid as a preconditioner (MGCG) features a good convergence even when the M...
详细信息
ISBN:
(纸本)9783642552243
Multigrid methods are among the fastest numerical algorithms for solving large sparse linear systems. the Conjugate Gradient method with Multigrid as a preconditioner (MGCG) features a good convergence even when the Multigrid solver itself is not efficient. the parallel FEM package NuscaS allows us to solve adaptive FEM problems with 3D unstructured meshes on parallel computers such as PC-clusters. the parallel version of the library is based on the geometric decomposition applied for computing nodes of a parallel system;the distributed-memory architecture and message-passing model of parallel programming are assumed. In our previous works, we extend the NuscaS functionality by introducing parallel adaptation of tetrahedral FEM meshes and dynamic load balancing capabilities. In this work we focus on efficient implementation of Geometric Multigrid as a parallel preconditioner for the Conjugate Gradient iterative solver used in the NuscaS package. Based on the geometric decomposition, for each level of Multigrid, meshes are partitioned and assigned to processors of a parallel architecture. Fine-grid levels are constructed by subdivision of mesh elements using the parallel 8-tetrahedra longestedge refinement mesh algorithm, where every process keeps the assigned part of mesh on each level of Multigrid. the efficiency of the proposed implementation is investigated experimentally.
Public Sensing (PS) is a recent trend for building large-scale sensor data acquisition systems using commodity smartphones. Limiting the energy drain on participating devices is a major challenge for PS, as otherwise ...
详细信息
ISBN:
(纸本)9783319115689
Public Sensing (PS) is a recent trend for building large-scale sensor data acquisition systems using commodity smartphones. Limiting the energy drain on participating devices is a major challenge for PS, as otherwise people will stop sharing their resources withthe PS system. Existing solutions for limiting the energy drain through model-driven optimizations are limited to dense networks where there is a high probability for every point of interest to be covered by a smartphone. In this work, we present an adaptive model-driven PS system that deals with both dense and sparse networks. Our evaluations show that this approach improves data quality by up to 41 percentage points while enabling the system to run with a greatly reduced number of participating smartphones. Furthermore, we can save up to 81% of energy for communication and sensing while providing data matching an error bound of 1 degrees C up to 96% of the time.
the proliferation of data acquisition devices like 3D laser scanners had led to the burst of large-scale spatial terrain data which imposes many challenges to spatial data analysis and computation. Withthe advent of ...
详细信息
the proliferation of data acquisition devices like 3D laser scanners had led to the burst of large-scale spatial terrain data which imposes many challenges to spatial data analysis and computation. Withthe advent of several emerging collaborative cloud technologies, a natural and cost-effective approach to managing such large-scale data is to store and share such datasets in a publicly hosted cloud service and process the data within the cloud itself using modern distributedcomputing paradigms such as MapReduce. For several key spatial data analysis and computation problems, polygon retrieval is a fundamental operation which is often computed under real-time constraints. However, existing sequential algorithms fail to meet this demand effectively given that terrain data in recent years have witnessed an unprecedented growth in both volume and rate. In this work, we develop a MapReduce-based parallel polygon retrieval algorithm which aims at minimizing the IO and CPU loads of the map and reduce tasks during spatial data processing. the results of the preliminary experiments on a Hadoop cluster demonstrate that the proposed techniques are scalable and lead to more than 35% reduction in execution time of the polygon retrieval operation over existing distributed algorithms.
We consider an emerging class of challenging networked applications called Real-Time Online Interactive applications (ROIA). ROIA are networked applications connecting a potentially very high number of users who inter...
We consider an emerging class of challenging networked applications called Real-Time Online Interactive applications (ROIA). ROIA are networked applications connecting a potentially very high number of users who interact withthe application and with each other in real time, i.e., a response to a user's action happens virtually immediately. Typical representatives of ROIA are multiplayer online computer games, advanced simulation-based e-learning and serious gaming. All these applications are characterized by high performance and QoS requirements, such as: short response times to user inputs (about 0.1-1.5 s); frequent state updates (up to 100 Hz); large and frequently changing numbers of users in a single application instance (up to tens of thousands simultaneous users).
parallel pattern libraries (e.g., Intel TBB) are popular and useful tools for developing applications in SMP environments at a higher level of abstraction. Such libraries execute user-provided code efficiently on shar...
详细信息
parallel pattern libraries (e.g., Intel TBB) are popular and useful tools for developing applications in SMP environments at a higher level of abstraction. Such libraries execute user-provided code efficiently on shared memory parallel architectures in accordance with well-defined execution patterns like parallel for-loops or pipelines. For heterogeneous architectures comprised of CPUs and accelerators, OpenCL has gained a lot of momentum. Since accelerated architectures do not provide a shared memory, it is not possible to directly use the approach taken in pattern libraries for SMP systems for OpenCL as well. In this paper, we are exploring issues and opportunities encountered by attempts to provide such patterns in the context of OpenCL. Based on a set of experiments with a scientific application on diverse OpenCL devices, we point out major pitfalls and insights, and outline directions for further efforts in developing pattern libraries for OpenCL.
Withthe increasing amount of digital image data, massive image process and feature extraction process have become a time-consuming process. As an excellent mass data processing and storage capacity of the open source...
详细信息
ISBN:
(纸本)9783319093338;9783319093321
Withthe increasing amount of digital image data, massive image process and feature extraction process have become a time-consuming process. As an excellent mass data processing and storage capacity of the open source cloud platform, Hadoop provides a parallelcomputing model MapReduce, HDFS distributed file system module. Firstly, we introduced Hadoop platform programming framework and Tamura texture features. And then, the image processing and feature texture feature extraction calculations involved in the process to achieve Hadoop platform. the results which comparison with Matlab platform shows it is less obvious advantage of Hadoop platform in image processing and feature extraction of lower-resolution images, but for image processing and feature extraction of high-resolution images, the time spent in Hadoop platform is greatly reducing, data processing capability the advantages is obvious.
暂无评论