Flexible and Realistic Client-centric Evaluation of Personal Clouds via User Stereotypes

Identifier:  TFM:156

Handle: https://hdl.handle.net/20.500.11797/TFM156

Authors:  Chenglong Zou

Abstract:
Abstract The performance evaluation of Personal Clouds (e.g., Dropbox, Google Drive, Box) is challenging due to their fat-client architecture and the absence of realistic workload generators that reproduce the disparate types of user behavior. In this work, we propose a novel storage benchmarking methodology in which the workload is generated by emulating groups of users that exhibit a similar behavior, namely, user stereotypes. Weinstantiate this methodology in BenchBox, a distributed performance analysis tool that integrates: (i) A practical yet accurate model that locally reproduces the aspects of user behavior that matter to the performance of desktop clients (e.g., types of activity, data contents); (ii) A centralized management and real-time monitoring system of Personal Cloud desktop clients. (iii) BenchBox also uses compact files to represent user stereotypes, enabling researchers to share only the behavior of users withoutdisclosing private traces. We show the benefits of BenchBox by exercising public and private Personal Clouds. We show how distinct types of users impact on the performance and efficiency of these systems, which may guide their optimization. Keywords: Personal Cloud Storage, Performance Analysis, Workload Generator, Stereotype.