Considering common mode failure (CMF) in the rendering cluster systems, the availability of rendering cluster systems with the increase of cluster’s number was studied. Firstly, based on availability of system with one cluster node, system with two cluster nodes was modeled with continuous time markov chain (CTMC) model. Then, the CTMC model was extended to the case of system with three cluster nodes. Furthermore, by solving these three CTMC models, availability for different cases were numerically deduced. Additionally, during one year the unavailable time for different cases was calculated and analysis by comparison was conducted. Finally, conclusions on different cases’ advantages and disadvantages are derived thereby, which offers theoretical foundations for establishing rendering cluster systems.
Based on the characteristics of rendering phase, a performance-based policy for job assignment under Distributed Rendering Environments (DREs) is proposed. Firstly, a method of evaluating rendering nodes is designed by taking CPU ratio, RAM size, CPU occupation, RAM usage into account. Furthermore, the corresponding algorithm of job assignment for dividing rendering tasks into sub-tasks on available rendering nodes is presented. Whats more, this algorithms correctness is validated by three groups of experiments thereby which offers quantitative guide for optimal job assignment of rendering.
Dynamic evolving characteristic is the key of modeling internetware architecture. However, it is lack of formal description and theoretical analysis for model dynamic evolution for the moment, a new kind of internetware dynamic evolving modeling method based on Seal-Calculus of process algebra is presented and formalized analysis is conducted for evolving process in practical applications, which is convenient to express obviously and strictly the dynamic process of the system, providing a new means and theoretical basis.
In order to enhance the operation efficiency of RSA algorithm, a new improved algorithm was suggested in this paper which made some improvements in structure and operation, and it was applied to digital signature. The experiment made comparison between a combinatorial optimization algorithm which combined SMM with index of 2k hexadecimal algorithm and the new algorithm. It shows that the new algorithm reaches a high level in operation speed.
On the basis of summarizing the concept filtering methods in the current Ontology learning, a method of domain concept filtering in the semantic level based on combination of word embedding and conventional statistics was presented, which can identify low-frequency words well, and as far as possible to ensure universality. Through experimental contrast, the proposed approach was proved to have a higher accuracy rate than the ways based on statistics.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.