Search

MPI derived datatypes allow users to describe noncontiguous memory layout and communicate noncontiguous data with a single communication function. This powerful feature enables an MPI implementation to optimize the transfer of noncontiguous data. In practice, however, many implementations of MPI derived datatypes perform poorly, which makes application developers avoid using this feature. In this paper, we present a technique to automatically select templates that are optimized for memory performance based on the access pattern of derived datatypes. We implement this mechanism in the MPICH2 source code. The performance of our implementation is compared to well-written manual packing/unpacking routines and original MPICH2 implementation. We show that performance for various derived datatypes is significantly improved and comparable to that of optimized manual routines. Keywords: MPI, derived datatypes, MPI performance optimization. [ABSTRACT FROM AUTHOR]

With the rapid development of computer multimedia, how to protect digital products from being copied, pirated and juggled has been an urgent issue in the information security field. Digital watermarking is a new method to solve the referred problems. This paper proposes a digital watermarking algorithm for three-dimension (3D) mesh models. The watermark is embedded into the host with modifying the lengths between the vertices and the centroid of models. The embedded watermark is invisible and can withstand the common attacks such as polygon mesh simplifications, addition of random noise, and similarity transforms. The validity of proposed algorithm has been confirmed with experiments. [ABSTRACT FROM AUTHOR]

This paper proposes a semi-fragile speech watermarking scheme by quantization of Linear Prediction (LP) parameters, i.e. the inverse sine parameters. The watermark decoding performance is analyzed by modelling the parameters estimation error as Laplace distributed noises. The watermark detection threshold is derived according to the requirement of error probability and expected Signal to Noise Ratio (SNR). Experiments show that the proposed watermarking scheme is robust against amplitude scaling and semi-fragile to white noise addition, and thus suitable for speech authentication. [ABSTRACT FROM AUTHOR]

The mean and standard deviation of the magnitudes of coefficients of the transform domain are common used in texture image retrieval. This paper performs an experimental comparison on Gabor wavelet and wavelet frame based features for this application. The detail comparison results indicate the performance of the former is not superior over that of the last in the query-by-example retrieval. [ABSTRACT FROM AUTHOR]

In this paper, a novel application of SVG (Scalable Vector Graphics) is proposed in watermarking. Vector graphic is used to detect geometric transform and control geometric distortion. Experimental results show that the proposed method is efficient to solve geometric attack to watermarking. [ABSTRACT FROM AUTHOR]

In this paper, a novel watermarking scheme in the compressed domain is proposed for video copyright protection. The robust watermark is embedded in motion vectors of the bitstream based on the relationships between one-pixel accuracy and half-pixel accuracy motion vectors at the encoder. Watermark can be blindly extracted by parsing the properties of motion vectors at the decoder. Simulation results show that the proposed algorithm can perform the real-time watermark embedding process without affecting visual quality of the video sequence and it is also standard-compliant. The proposed algorithm can be applied to any video compression standard with half-pixel accuracy motion vectors such as MPEG-2 and H.263. [ABSTRACT FROM AUTHOR]

Task Scheduling is a critical design issue of distributed computing. The emerging Grid computing infrastructure consists of heterogeneous resources in widely distributed autonomous domains and makes task scheduling even more challenging. Grid considers both static, unmovable hardware and moveable, replicable data as computing resources. While intensive research has been done on task scheduling on hardware computing resources and on data replication protocols, how to incorporate data movement into task scheduling seamlessly is unrevealed. We consider data movement as a dimension of task scheduling. A dynamic data structure, Data Distance Table (DDT), is proposed to provide real-time data distribution and communication information. Based on DDT, a data-conscious task scheduling heuristics is introduced to minimize the data access delay. A simulated Grid environment is set up to test the efficiency of the newly proposed algorithm. Experimental results show that for data intensive tasks, the dynamic data-conscious scheduling outperforms the conventional Min-Min significantly. [ABSTRACT FROM AUTHOR]