Heuristic file sorted assignment algorithm of parallel I/O on cluster computing system

A new file assignment strategy of parallel I/O, which is named heuristic file sorted assignment algorithm was proposed on cluster computing system. Based on the load balancing, it assigns the files to the same disk according to the similar service time. Firstly, the files were sorted and stored at the set I in descending order in terms of their service time, then one disk of cluster node was selected randomly when the files were to be assigned, and at last the continuous files were taken orderly from the set I to the disk until the disk reached its load maximum. The experimental results show that the new strategy improves the performance by 20.2% when the load of the system is light and by 31.6% when the load is heavy. And the higher the data access rate, the more evident the improvement of the performance obtained by the heuristic file sorted assignment algorithm.

Parallel Computation of Shallow-water Model on Workstations Cluster

ＰａｒａｌｌｅｌＣｏｍｐｕｔａｔｉｏｎｏｆＳｈａｌｌｏｗ－ｗａｔｅｒＭｏｄｅｌｏｎＷｏｒｋｓｔａｔｉｏｎｓＣｌｕｓｔｅｒＳｏｎｇＪｕｎｑｉａｎｇ；ＳｕｎＡｎ－ｃｌａｎｇ，；ＬｉＸｉａｏｍｅｉ（ｅｐａｒｔｍｅｎｔＯｆＣＯ，ｍｐ＇ｕｔｅｒＳｃｉｅｎｃｅ，...

Energy Optimization Oriented Three-Way Clustering Algorithm for Cloud Tasks

Cloud computing has developed as an important information technology paradigm which can provide on-demand services. Meanwhile,its energy consumption problem has attracted a grow-ing attention both from academic and industrial communities. In this paper,from the perspective of cloud tasks,the relationship between cloud tasks and cloud platform energy consumption is established and analyzed on the basis of the multidimensional attributes of cloud tasks. Furthermore,a three-way clustering algorithm of cloud tasks is proposed for saving energy. In the algorithm,f irst,t he cloud tasks are classified into three categories according to the content properties of the cloud tasks and resources respectively. Next,cloud tasks and cloud resources are clustered according to their computation characteristics（ e. g. computation-intensive,data-intensive）. Subsequently,greedy scheduling is performed. The simulation results showthat the proposed algorithm can significantly reduce the energy cost and improve resources utilization,compared with the general greedy scheduling algorithm.

An Online Visualization System for Streaming Log Data of Computing Clusters

Monitoring a computing cluster requires collecting and understanding log data generated at the core, computer, and cluster levels at run time. Visualizing the log data of a computing cluster is a challenging problem due to the complexity of the underlying dataset： it is streaming, hierarchical, heterogeneous, and multi-sourced. This paper presents an integrated visualization system that employs a two-stage streaming process mode. Prior to the visual display of the multi-sourced information, the data generated from the clusters is gathered, cleaned, and modeled within a data processor. The visualization supported by a visual computing processor consists of a set of multivariate and time variant visualization techniques, including time sequence chart, treemap, and parallel coordinates. Novel techniques to illustrate the time tendency and abnormal status are also introduced. We demonstrate the effectiveness and scalability of the proposed system framework on a commodity cloud-computing platform.

Hybrid Decomposition Method in Parallel Molecular Dynamics Simulation Based on SMP Cluster Architecture

A hybrid decomposition method for molecular dynamics simulations was presented, using simul- taneously spatial decomposition and force decomposition to fit the architecture of a cluster of symmetric multi-processor (SMP) nodes. The method distributes particles between nodes based on the spatial decom- position strategy to reduce inter-node communication costs. The method also partitions particle pairs within each node using the force decomposition strategy to improve the load balance for each node. Simulation results for a nucleation process with 4 000 000 particles show that the hybrid method achieves better paral- lel performance than either spatial or force decomposition alone, especially when applied to a large scale particle system with non-uniform spatial density.

Towards application-level elasticity on shared cluster： an actor-based approach

In current cluster computing, several distributed frameworks are designed to support elasticity for business services adapting to environment fluctuation. However, most existing works support elasticity mainly at the resource level, leaving application level elasticity support problem to domain-specific frameworks and applications. This paper proposes an actor-based general approach to support application-level elasticity for multiple cluster computing frameworks. The actor model offers scalability and decouples language-level concurrency from the runtime environment. By extending actors, a new middle layer called Unisupervisor is designed to ＂sit＂ between the resource management layer and application framework layer. Actors in Unisupervisor can automatically distribute and execute tasks over clusters and dynamically scale in/out. Based on Unisupervisor, high-level profiles （MasterSlave, MapReduce, Streaming, Graph, and Pipeline） for diverse cluster computing requirements can be supported. The entire approach is implemented in a prototype system called UniAS. In the evaluation, both benchmarks and real applications are tested and analyzed in a small scale cluster. Results show that UniAS is expressive and efficiently elastic.

High-Performance Broadcasting Algorithms on Cluster

In many clusters connected by high-speed communication networks, the exact structure of the underlying communication network and the latency difference between different sending and receiving pairs may be ignored when they broadcast, such as in the approach adopted by the broadcasting method in MPICH, a widely used MPI implementation. However, the underlying network cluster topologies are becoming more and more complicated and the performance of traditional broadcasting algorithms, such as MPICHs MPI_Bcast, is far from good. This paper analyzed the impact of communication latencies and the underlying topologies on the performance of broadcasting algorithms for multilevel clusters. A multilevel model was developed for broadcasting in clusters with complicated topologies, which divides the cluster topology into many levels based on the underlying topology. The multilevel model was used to develop a new broadcast algorithm, MLM broadcast-2 (MLMB-2), that adapts to a wide range of clusters. Comparison of the performance of the counterpart MPI operation MPI_Bcast and MLMB-2 shows that MLMB-2 outperforms MPI_Bcast by decreasing the broadcast running time by 60%-90%.

A novel fragile watermarking scheme for image tamper detection and recovery

We propose a fragile watermarking scheme capable of image tamper wise dependency mechanism. Initially, the image is divided into detection and recovery with a block blocks with size improve image tamper localization precision. By combining image local properties of 2~2 in order to with human visual system, authentication data are acquired. By computing the class membership degree of each image block property, data are generated by applying k-mean clustering technique to cluster all image blocks. The recovery data are composed of average intensity obtained by truncating the two least significant bits （LSBs） of each pixel within each block. Finally, the logistic chaotic encrypted feature watermark consisting of 2-bit authentication data and 6-bit recovery data of image block is embedded into the two LSBs of each pixel within its corresponding mapping block. Experimental results show that the proposed algorithm does not only achieve superior tamper detection and locate tiny tampered positions in images accurately, it also recovers tampered regions effectively.

On/Off-Line Prediction Applied to Job Scheduling on Non-Dedicated NOWs

This paper proposes a prediction engine designed for non-dedicated clusters, which is able to estimate the turnaround time for parallel applications, even in the presence of serial workload of the workstation owner. The prediction engine can be configured to work with three different estimation kernels： a Historical kernel, a Simulation kernel based on analytical models and an integration of both, named Hybrid kernel. These estimation proposals were integrated into a scheduling system, named CISNE, which can be executed in an on-line or off-line mode. The accuracy of the proposed estimation methods was evaluated in relation to different job scheduling policies in a real and a simulated cluster environment. In both environments, we observed that the Hybrid system gives the best results because it combines the ability of a simulation engine to capture the dynamism of a non-dedicated environment together with the accuracy of the historical methods to estimate the application runtime considering the state of the resources.