Web proxy cache replacement scheme based on back-propagation neural network

Web proxy caches are used to reduce the strain of contemporary web traffic on web servers and network bandwidth providers. In this research, a novel approach to web proxy cache replacement which utilizes neural networks for replacement decisions is developed and analyzed. Neural networks are trained to classify cacheable objects from real world data sets using information known to be important in web proxy caching, such as frequency and recency. Correct classification ratios between 0.85 and 0.88 are obtained both for data used for training and data not used for training. Our approach is compared with Least Recently Used (LRU), Least Frequently Used (LFU) and the optimal case which always rates an object with the number of future requests. Performance is evaluated in simulation for various neural network structures and cache conditions. The final neural networks achieve hit rates that are 86.60% of the optimal in the worst case and 100% of the optimal in the best case. Byte-hit rates are 93.36% of the optimal in the worst case and 99.92% of the optimal in the best case. We examine the input-to-output mappings of individual neural networks and analyze the resulting caching strategy with respect to specific cache conditions.

[1]  Luigi Rizzo,et al.  Replacement policies for a proxy cache , 2000, TNET.

[2]  Sang Lyul Min,et al.  Efficient Replacement of Nonuniform Objects in Web Caches , 2002, Computer.

[3]  Jacques de Villiers,et al.  Backpropagation neural nets with one and two hidden layers , 1993, IEEE Trans. Neural Networks.

[4]  Martin F. Arlitt,et al.  Evaluating content management techniques for Web proxy caches , 2000, PERV.

[5]  R. Lippmann,et al.  An introduction to computing with neural nets , 1987, IEEE ASSP Magazine.

[6]  Robert M. Farber,et al.  How Neural Nets Work , 1987, NIPS.

[7]  Anja Feldmann,et al.  Web proxy caching: the devil is in the details , 1998, PERV.

[8]  Hu Yuqi,et al.  Caching on the World Wide Web , 2003 .

[9]  Sandy Irani,et al.  Cost-Aware WWW Proxy Caching Algorithms , 1997, USENIX Symposium on Internet Technologies and Systems.

[10]  Humayun Khalid.  A new cache replacement scheme based on backpropagation neural networks , 1997, CARN.

[11]  Patrick van der Smagt.  A Comparative Study of Neural Network Algorithms Applied to Optical Character Recognition , 1990, IEA/AIE.

[12]  László Böszörményi,et al.  A survey of Web cache replacement strategies , 2003, CSUR.

[13]  LiMin Fu,et al.  Knowledge discovery based on neural networks , 1999, Commun. ACM.

[14]  Hava T. Siegelmann,et al.  On a learnability question associated to neural networks with continuous activations (extended abstract) , 1994, COLT '94.

[15]  Eric A. Brewer,et al.  Value-based web caching , 2003, WWW '03.

[16]  Jake Cobb.  USING NEURAL NETWORKS FOR WEB PROXY CACHE REPLACEMENT , 2005 .

[17]  Hala ElAarag,et al.  TRAINING AND SIMULATION OF NEURAL NETWORKS FOR WEB PROXY CACHE REPLACEMENT , 2006 .

[18]  Humayun Khalid.  Performance of the KORA-2 cache replacement scheme , 1997, CARN.