In virtual MIMO technology, distributed singleantenna radio systems cooperate on information transmission and reception as a multiple- antenna MIMO radio system. In this paper, a new virtual MIMO communication scheme ...
详细信息
To use software by accessing it across the Internet instead of installing and using it on-premises becomes more and more common. This model is called software as a Service. It allows the customers to concentrate on th...
详细信息
ISBN:
(纸本)9781617386077
To use software by accessing it across the Internet instead of installing and using it on-premises becomes more and more common. This model is called software as a Service. It allows the customers to concentrate on their core business and to save money due to less operational costs. Hence, many software vendors are nowadays in need to modernize their existing application for a SaaS offering to catch up with the market. In doing so, the vendor can pursue different approaches, especially considering how to enable the application for multi-tenancy that constitutes a key factor for a higher profit by leveraging economies of scale. We introduce three engineering approaches to modernize a data-intensive enterprise application for a multi-tenant SaaS offering: Reuse, Recycle, and Redesign. Further, we identify different engineering issues and isolation issues between tenants with respect to the approach Recycle. Finally, we contribute an initial strategy for recycling data-intensive enterprise applications.
The increases in multi-core processor parallelism and in the flexibility of many-core accelerator processors, such as GPUs, have turned traditional SMP systems into hierarchical, heterogeneous computing environments. ...
详细信息
ISBN:
(纸本)9781605589428
The increases in multi-core processor parallelism and in the flexibility of many-core accelerator processors, such as GPUs, have turned traditional SMP systems into hierarchical, heterogeneous computing environments. Fully exploiting these improvements in parallel system design remains an open problem. Moreover, most of the current tools for the development of parallel applications for hierarchical systems concentrate on the use of only a single processor type (e.g., accelerators) and do not coordinate several heterogeneous processors. Here, we show that making use of all of the heterogeneous computing resources can significantly improve application performance. Our approach, which consists of optimizing applications at run-time by efficiently coordinating application task execution on all available processing units is evaluated in the context of replicated dataflow applications. The proposed techniques were developed and implemented in an integrated run-time system targeting both intra- and inter-node parallelism. The experimental results with a real-world complex biomedical application show that our approach nearly doubles the performance of the GPU-only implementation on a distributed heterogeneous accelerator cluster. Copyright 2010 ACM.
Current distributedsystems are usually composed of several distributed components that communicate through specific ports. When testing these systems we separately observe sequences of inputs and outputs at each port...
详细信息
software applications are increasingly situated in a world where context changes continuously. At the same time, applications need to provide continuous service, and the service provided often needs to change in order...
详细信息
Binding site mapping refers to the computational prediction of the regions on a protein surface that are likely to bind a small molecule with high affinity. The process involves flexibly docking a variety of small mol...
详细信息
作者：
Ranok, UdomKittitornkun, SurinTongsima, SissadesFaculty of Engineering
King Mongkut's Institute of Technology Lardkrabang School of Computer Engineering and Information Science Bangkok 10520 Thailand Genome Institute
National Center for Genetic Engineering and Biotechnology 113 Thailand Science Park Paholyothin Road Klong 1 Klong Luang Pathumtani 12120 Thailand
This paper presents a parallelization framework for infer- ring haplotypes using an expectation maximization (EM) algorithm. Our framework utilizes GProf profiling tool, OpenMP library, and ompP profiling tool to para...
详细信息
ISBN:
(纸本)9780889868786
This paper presents a parallelization framework for infer- ring haplotypes using an expectation maximization (EM) algorithm. Our framework utilizes GProf profiling tool, OpenMP library, and ompP profiling tool to parallelize the algorithm by determining the hotspot functions, mul- tithreading, and executing them on the Multi-core CPUs. In our experiments, we choose the SNPHAP program for this case study and run it on an 8-core Xeon Linux ma- chine. The results show that our framework can signifi- cantly speedup up to 214% on a large data set with 151 loci of a 10,000 data samples. In addition, deep profiles of multithreaded SNPHAP support our discovery that maxi- mum speedup can be achieved when the number of parallel threads equals to the number of physical cores.
This paper describes the development of an acoustic distributed beamforming system and presents experimental results for two-source and three-source acoustic distributed beamforming using the time-slotted round-trip c...
详细信息
We discuss the initial phases of software development of a real industrial safety-related device in the railway application domain. In particular, to achieve greater confidence in the system, we illustrate the develop...
详细信息
A graph G is called k-fault Hamiltonian (resp. Hamiltonian-connected) if after deleting at most k vertices and/or edges from G, the resulting graph remains Hamiltonian (resp. Hamiltonian-connected). Let δ(G) be the m...
详细信息
暂无评论