Web Workloads: how things have changed and what has remained the same

Virgilio Almeida - UFMG/Brasil

Abstract

The design of Web applications and services require good models of the workloads to which these applications are subject. The Web is a complex system with hundreds of millions of users demanding services from thousands of applications. In order to understand the Web, one needs first to understand the workload associated with their most popular applications, such as email, peer-to-peer, e-business, search engines, streaming media, and social networks. This talk presents an overview of a variety of real Web workloads and how they evolved in the last ten years. It shows the main properties of these workloads and discusses the invariants across different types of workloads. It introduces methodologies and techniques used in workload characterization and modeling. Constructing a model involves tradeoffs between realism and complexity. The talk shows how characterization techniques have been used to capture the most relevant aspects of Web workloads while keeping the model as simple as possible. The talk also discusses how workload models can be used to design efficient Web systems and services.

Short Bio

Virgilio Almeida is a professor of the Computer Science Department at the Federal University of Minas Gerais, Brazil. His research interests include performance modeling and analysis of large scale distributed systems and large social networks. In particular, his current research work is focused on the interaction of social networks and system behavior, including factors such as performance, availability and malicious behavior.

Virgilio is a recipient of a Fulbright Research Scholar Award. He held visiting positions at Boston University, Polytechnic University of Catalunya in Barcelona, Polytechnic University-Brooklyn and held visiting appointments at Xerox PARC and Hewlett-Packard Research Laboratory. Professor Almeida is co-author of 4 books on performance modeling (i.e., "Performance by Design: computer capacity planning by example," "Capacity Planning for Web Services: metrics, models, and methods," ) and is the author of over eighty papers on several topics in computer systems.