Cluster计算能力模型总结与设想  

An Evaluation & Assumption on Performance of Cluster Computing

作　　者：刘乃维[1] 

机构地区：[1]北京航空航天大学,北京100191

出　　处：《电脑开发与应用》2013年第12期70-72,共3页Computer Development & Applications

摘　　要：Cluster计算,中文名称为集群计算。它可以被认为是这样一种技术:它把许多系统连接到一起,使多台服务器像一台机器一样进行计算工作。采用cluster计算,通常是为了提高系统稳定性的同时,提高系统对大数据的实时处理能力。在cluster计算中,服务器群在网络中表现为单一的系统,并以这种模式加以管理。对于并行计算和集群计算,现已提出的各种计算模型,大多针对的是并行计算与它的共性,而非特性。而如果需要针对这些特性,就需要用一种更加准确的模型加以描述。将介绍基于集群的一些计算模型,并提出新的cluster计算模型的一个可能的方案。Cluster computing, known as a kind of computing technology to connect the systems, letting them work like a single processor,using cluster computing in dealing with big data could help the computing system performance better and more powerful in security at the same time. This multi-processor system, shown as a single system in the network, alsois managed in this way. There are many models to analyze the performance of cluster computing, which aimed just at the common point it shared with parallel computing. In fact, this paper need to find a way to analyze and model the performance of cluster computing based on the feature of its own. The models used are evaluated now, and then propose a better way to analyze.

关 键 词：集群 并行计算模型 超级步 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]