Fluid limit analysis of FIFO and RR caching for independent reference models

We study the fluid limit analysis of random replacement (RR) caching for the independent reference model. Applying the limit theorem for the mean field interaction model, we derive the fluid limit of fault probability in the transient state as well as in the steady state. Since the stationary fault probability for the RR cache is identical to that for the first-in first-out (FIFO) cache, our results on the stationary fault probability are available for the FIFO caching. We see that the fluid limit of stationary fault probability that we obtain is coincident with the existing heuristic approximation of stationary fault probability. That is, our fluid limit analysis gives a rigorous theoretical foundation to the heuristic approximation.

[1]  Erol Gelenbe,et al.  A Unified Approach to the Evaluation of a Class of Replacement Algorithms , 1973, IEEE Transactions on Computers.

[2]  Takayuki Osogami,et al.  Analysis of Page Replacement Policies in the Fluid Limit , 2010, Oper. Res..

[3]  Asit Dan,et al.  An approximate analysis of the LRU and FIFO buffer replacement schemes , 1990, SIGMETRICS '90.

[4]  Dennis Shasha,et al.  2Q: A Low Overhead High Performance Buffer Management Replacement Algorithm , 1994, VLDB.

[5]  Jean-Yves Le Boudec,et al.  A class of mean field interaction models for computer and communication systems , 2008, Perform. Evaluation.

[6]  Alberto Gandolfi,et al.  LRU is better than FIFO under the independent reference model , 1992, Journal of Applied Probability.

[7]  P. Jelenkovic.  Asymptotic approximation of the move-to-front search cost distribution and least-recently used caching fault probabilities , 1999 .

[8]  Yuanyuan Zhou,et al.  The Multi-Queue Replacement Algorithm for Second Level Buffer Caches , 2001, USENIX Annual Technical Conference, General Track.

[9]  Kumiko Hattori,et al.  Hydrodynamic limit of move-to-front rules and search cost probabilities , 2009, ArXiv.

[10]  Alexandre Proutière,et al.  A particle system in interaction with a rapidly varying environment: Mean field limits and applications , 2010, Networks Heterog. Media.

[11]  Jean-Yves Le Boudec,et al.  The stationary behaviour of fluid limits of reversible processes is concentrated on stationary points , 2010, Networks Heterog. Media.