Optimization inWeb Caching: Cache Management, Capacity Planning, and Content Naming

Caching;Web Cache Hierarchies;Computer Science and Engineering

Kelly, Terence P.Gray, Jim ;

University of Michigan

关键词: Caching;    Web Cache Hierarchies;    Computer Science and Engineering;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/90029/kelly-optimization_web_caching.ps.bz2?sequence=2&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Caching is fundamental to performance in distributed information retrieval systemssuch as the World Wide Web. This thesis introduces novel techniques for optimizing performanceand cost-effectiveness in Web cache hierarchies.When requests are served by nearby caches rather than distant servers, server loads andnetwork traffic decrease and transactions are faster. Cache system design and management,however, face extraordinary challenges in loosely-organized environments like the Web,where the many components involved in content creation, transport, and consumption areowned and administered by different entities. Such environments call for decentralizedalgorithms in which stakeholders act on local information and private preferences.In this thesis I consider problems of optimally designing new Web cache hierarchiesand optimizing existing ones. The methods I introduce span the Web from point of contentcreation to point of consumption: I quantify the impact of content-naming practices oncache performance; present techniques for variable-quality-of-service cache management;describe how a decentralized algorithm can compute economically-optimal cache sizes ina branching two-level cache hierarchy; and introduce a new protocol extension that eliminatesredundant data transfers and allows ;;dynamic” content to be cached consistently.To evaluate several of my new methods, I conducted trace-driven simulations on anunprecedented scale. This in turn required novel workload measurement methods and efficientnew characterization and simulation techniques. The performance benefits of my proposedprotocol extension are evaluated using two extraordinarily large and detailed workloadtraces collected in a traditional corporate network environment and an unconventionalthin-client system.My empirical research follows a simple but powerful paradigm: measure on a largescale an important production environment’s exogenous workload; identify performancebounds inherent in the workload, independent of the system currently serving it; identifygaps between actual and potential performance in the environment under study; and finallydevise ways to close these gaps through component modifications or through improvedinter-component integration. This approach may be applicable to a wide range of Webservices as they mature.