Characterizing reference locality in the WWW Export

@inproceedings{citeulike:3482055, abstract = {In this paper we propose models for both temporal and spatial locality of reference in streams of request arriving at Web servers. We show that simple models based on document popularity alone are insufficient for capturing either temporal or spatial locality. Instead, we rely on an equivalent, but numerical, representation of a reference stream: a stack distance trace. We show that temporal locality can be characterized by the marginal distribution of the stack distance trace, and we propose models for typical distributions and compare their cache performance to our traces. We also show that spatial locality in a reference stream can be characterized using the notion of self-similarity. Self-similarity describes long-range correlations in the dataset, which is a property that previous researchers have found hard to incorporate into synthetic reference strings. We show that stack distance strings appear to be strongly self-similar, and we provide measurements of the degree of self-similarity in our traces. Finally, we discuss methods for generating synthetic Web traces that exhibit the properties of temporal and spatial locality that we measured in our data.}, address = {Washington, DC, USA}, author = {Almeida, Virg\'\ilio and Bestavros, Azer and Crovella, Mark and de Oliveira, Adriana}, booktitle = {DIS '96: Proceedings of the fourth international conference on on Parallel and distributed information systems}, citeulike-article-id = {3482055}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=382006.383200}, isbn = {0-8186-7475-X}, keywords = {access-pattern, distribution, irm, web, zipf}, location = {Miami Beach, Florida, United States}, pages = {92--107}, posted-at = {2008-11-05 11:03:33}, priority = {2}, publisher = {IEEE Computer Society}, title = {Characterizing reference locality in the WWW}, url = {http://portal.acm.org/citation.cfm?id=382006.383200}, year = {1996} }

TY - CONF ID - citeulike:3482055 L3 - citeulike-article-id:3482055 N2 - In this paper we propose models for both temporal and spatial locality of reference in streams of request arriving at Web servers. We show that simple models based on document popularity alone are insufficient for capturing either temporal or spatial locality. Instead, we rely on an equivalent, but numerical, representation of a reference stream: a stack distance trace. We show that temporal locality can be characterized by the marginal distribution of the stack distance trace, and we propose models for typical distributions and compare their cache performance to our traces. We also show that spatial locality in a reference stream can be characterized using the notion of self-similarity. Self-similarity describes long-range correlations in the dataset, which is a property that previous researchers have found hard to incorporate into synthetic reference strings. We show that stack distance strings appear to be strongly self-similar, and we provide measurements of the degree of self-similarity in our traces. Finally, we discuss methods for generating synthetic Web traces that exhibit the properties of temporal and spatial locality that we measured in our data. CY - Miami Beach, Florida, United States SN - 0-8186-7475-X AD - Washington, DC, USA EP - 107 TI - Characterizing reference locality in the WWW PB - IEEE Computer Society T2 - DIS '96: Proceedings of the fourth international conference on on Parallel and distributed information systems SP - 92 KW - access-pattern KW - distribution KW - irm KW - web KW - zipf AU - Almeida, Virg\'\ilio AU - Bestavros, Azer AU - Crovella, Mark AU - de Oliveira, Adriana PY - 1996/// UR - http://portal.acm.org/citation.cfm?id=382006.383200 ER -