User behavior models are important for building realistic simulation environment for research on P2P multimedia file-sharing systems. In this paper, we build a user download behavior model and a user removal behavior model, which can describe important user characteristics that are not captured in the existing models. The proposed download behavior model incorporates retry behavior, and the removal behavior model integrates free-riding, file usage, and file removal. Based on two-year real user logs, we derive the range of all the model parameters and generate many interesting observations. To validate the proposed models, we compare several models in a case study on the number of living replicas in P2P file-sharing systems. The results demonstrate the accuracy, usability, and advantage of the proposed models.Index Terms-Modeling, Multimedia systems
INTRODUCTIONFor the research on P2P multimedia file-sharing systems, new algorithms and schemes are mainly evaluated through simulations because experimental testing in real P2P networks is prohibitively expensive [1]. Therefore, building realistic simulation environment is critical.The major obstacle toward building the realistic simulation environment is the lack of a throughout understanding of dynamic behaviors of P2P users, which are affected by technology as well as human factors. The most important behaviors are file download and file removal. Most of the existing user behavior models, however, cannot capture two important factors in file download and file removal behaviors.• retry behavior: how users retry after download failure; • retention time: the time interval between a user downloading a file and removing it out of the P2P file-sharing system. It is affected by (i) free-riding, (ii) delay between downloading and examination of the files, and (iii) additional retention time after examination. The above two factors can greatly affect simulation results. However, it is challenging to understand and model these factors because it is hard to capture retry behavior and retention time directly from most of the current P2P systems. To our best knowledge, none of the existing work have parameterized these two factors from real user logs.In this paper, we build user behavior models based on two-year user logs in a real-world P2P network that is mainly used to exchange
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.