- From: Alex Lopez-Ortiz <alopez-o@barrow.uwaterloo.ca>
- Date: Wed, 7 Feb 1996 11:18:55 -0500 (EST)
- To: Z.Wang@cs.ucl.ac.uk (Zheng Wang)
- Cc: luigi@labinfo.iet.unipi.it, http-caching@pa.dec.com
> > > >There is a theorem in statistics which says that the the combination of > >large numbers of random variables yields to a gaussian distribution, > >independently of the distribution of each single variable. Similarly, > >when we have a national or continental cache, I expect the number of users > >to be so large to make differences disappear and make the aggregate > >result almost the same everywhere. > >A number of studies suggested a cache hit rate of 30-50% and >the results seem to be very consistent. See the paper >by Marc Abrams et al "Caching Proxies: Limitations and >Potentials" in proceedings of WWW4 for references. I would suggest as well J. Gwertzman, M. Seltzer. The Case for Geographical Push-Caching, in VINO: The 1994 Fall Harvest, Technical Report TR-34-94, Center for Research in Computing Technology, Harvard University, December, 1994. A. Bestavros, Demand-based Document Dissemination for the World-Wide Web. Technical Report 95-003, Computer Science Department, Boston University. February, 1995. A. Bestavros, R. L. Carter, M. E. Crovella, C. R. Cunha, A. Heddaya, S. A. Mirdad, Application-Level Document Caching in the Internet. Technical Report BU-CS-95-002, Computer Science Deparment, Boston University, Revised March, 1995. All of which are available on the Web... Here at Waterloo we have also modelled push-caching, and obtained a cache hit rate of ~35%, as percentage of traffic volume, as opposed to total number of hits. (This paper was submitted to WWW5). Cheers, Alex -- Alex Lopez-Ortiz alopez-o@neumann.UWaterloo.ca http://daisy.uwaterloo.ca/~alopez-o FAX (519)-885-1208 Department of Computer Science University of Waterloo Waterloo, Ontario N2L 3G1 Canada
Received on Wednesday, 7 February 1996 16:58:40 UTC