Minyong Shi

CTMC-Based Availability Analysis of Cluster System with Multiple Nodes

The availability of cluster system with multiple nodes was investigated by taking common mode failure (CMF) into account. Firstly, the availability of system with one cluster node was analyzed. Then on the basis of analysis, contraposing to the cluster system with two nodes and three nodes, four continuous time markov chain (CTMC) models in the case of No CMF and in the case of CMF were constructed respectively. Moreover, numerical comparison of steady availability for these two cases was conducted by solving related linear equations and the conclusion on the priority and inferiority comparison of different cases is derived thereby, which offers theoretical foundation for establishing cluster systems.

The prediction of character based on recurrent neural network language model

This paper mainly talks about the Recurrent Neural Network and introduces a more effective neural network model named LSTM. Then, the paper recommends a special language model based on Recurrent Neural Network. With the help of LSTM and RNN language models, program can predict the next character after a certain character. The main purpose of this paper is to compare the LSTM model with the standard RNN model and see their results in character prediction. So we can see the huge potential of Recurrent Neural Network Language Model in the field of character prediction.

SPN-Based Performance Evaluation for Data and Voice Traffic Hybrid System

On the basis of analyzing the characteristics of voice traffic, this paper constructed a Stochastic Petri Net (SPN) model for data and ON-OFF voice traffic hybrid system. Then, average time delay of the system was analyzed and model-based simulation was conducted with Stochastic Petri Net Package (SPNP) 6.0. Consequently, variation trends of burst time and idle time on average time delay are derived thereby. The methodology of modeling and simulation in this paper can be further used to analyze the performance of multimedia hybrid traffic system.

A SPN-Based Delay Analysis of LEO Satellite Networks

In this paper, we focus on the average time delay analysis of Low Earth Orbit (LEO) satellite networks. Firstly, a geometrical analysis of LEO satellite networks is given according to the characteristics of communication process. By taking the communication process of LEO satellite networks into account, average time delay is formulated by four parameters, i.e. average altitude of LEO satellite networks, uplink bandwidth, Inter-satellite links (ISLs) bandwidth and downlink bandwidth. Then, a Stochastic Petri Net (SPN) model is constructed. Accordingly, in order to represent the characteristics of average time delay, the arrival rate in the SPN model is mapped from the four above-mentioned parameters. Furthermore, the effects of these parameters on average time delay are also analyzed with Stochastic Petri Net Package (SPNP) 6.0. Because of its easiness and accuracy, the proposed approach has great benefit to the design and performance analysis of satellite networks.

Implementation of the protein sequence model based on ant colony optimization algorithm

The present paper concerns itself with a theoretical calculation study on implementation and optimization of protein sequence issue. Using ant colony optimization algorithm (ACO) to build the Two-Dimensional Hydrophobic-Hydrophilic (2DHP) model. There are three parts in this paper. Starting with a brief introduction of the ant colony optimization and 2DHP model, then we propose a method on how to construct such model based on ant colony optimization algorithm. We end the study with the conclusion that ant colony optimization algorithm for protein sequence theoretical calculation can be improved.

Analysis of film data based on Neo4j

Film data analysis is important for website to find relationship between film data. Other than normal relational database, a novel and popular graph database, Neo4j uses graph related concepts to describe data models, the data can easily be nodes, edges and their associated attributes. By focusing on film data, Neo4j-based analysis is conducted in this paper. Firstly, Neo4j and Cypher Query Language are introduced. Then Neo4j is applied to analyze the associations among key objects in film data which are directors, actors etc. Neo4j database is good at dealing with complex and multi-connection data, using Neo4j database to store and manage film data makes it convenient for film data analysis.

CTMC-based numerical analysis of cluster system with middle states in the case of common mode failure

By taking both CMF (Common Mode Failure) and middle states into account, numerical analysis of availability of cluster system is analyzed. Firstly, a CTMC (Continuous Time Markov Chain) model of cluster system containing three cluster nodes is constructed with failure rate (λ), repair rate (μ), CMF rate (7), reconfiguration rate (β) and reset rate (á) etc. Then the steady state solution of the CMTC model is concerned by solving the equations with the combination of several parameters. Subsequently, the effect of probability of being in automatic recovery mode (k) on availability of cluster system is analyzed and discussed via numerical values-based chart thereby. This work offers quantitative analysis of cluster systems availability.

CTMC-Based Availability Analysis of Multiple Cluster Systems with Common Mode Failure

By taking common mode failure (CMF) into account, the availability of multiple cluster systems is investigated. Firstly, a Markov model of system with three cluster nodes is constructed with failure rate (λ), repair rate (µ) and CMF rate (γ). Then, the availability of system is formulated with related parameters by solving related equations. Furthermore, derivative analysis of the formulas property is conducted. Finally, numerical results of systems availability are plotted and the corresponding parameters influence on systems availability is derived thereby. This work provides numerical foundations for designing big data-based platforms with high performance.

DTMC-Based Performance Evaluation of Animation Rendering Node

Animation rendering node is the key component of distributed animation rendering system, which offers rendering service. The performance of animation rendering node has great impact on users experience. This paper uses DTMC-based model checking methodology to evaluate animation rendering nodes performance. Based on the analysis of animation rendering nodes working process, the probability of steady-state is chosen as the performance index to study different parameters influence on it. Furthermore, trend that the probability of steady-state of animation rendering node varies with the effect of probability from Updating state to Waiting state and probability of self-waiting of Rendering state is derived thereby.

SPN model-based performance evaluation of job scheduling plans for distributed rendering environment

Based on the workflow of jobs under DRE (distributed rendering environment), mathematical modeling and performance evaluation were used to investigate different job scheduling plans. Firstly, a SPN (Stochastic Petri Nets) model for job scheduling with different priorities was constructed. Furthermore, using methodologies of SPN model and queuing one, effect of rendering rate, job priority on average completion time was conducted with exemplification, which offers key quantity foundations for designing and choosing appropriate job scheduling algorithms.