论文部分内容阅读
静态分析模型能在负载固定的情况下预期不同因素对缓存算法性能的影响,为解决该模型在动态负载下预测精度不高的问题,结合实际用户访问行为研究,采用负载拟合的方法对此问题进行探讨,并提出了一种针对变动负载的间隔缓存类算法的动态性能模型。该模型可更准确估算系统实际性能,从而为用户控制等策略提供参考。实验结果表明,静态模型的缓存命中率预测结果比实际高70%以上,而该模型则能适应负载的变动,预测结果与实际结果差别在10%左右。
The static analysis model can predict the impact of different factors on the performance of the caching algorithm under the condition of fixed load. To solve the problem of low prediction accuracy of the model under dynamic load, combined with the actual user access behavior, This paper also presents a dynamic performance model of interval buffer class algorithm for variable load. The model can estimate the actual performance of the system more accurately and provide reference for strategies such as user control. Experimental results show that the static model cache hit rate prediction results more than 70% higher than the actual, while the model is able to adapt to load changes, the prediction results and the actual results of about 10% difference.