This paper presents a fluid model and performance analysis of BitTorrent, a second-generation peer-to-peer (P2P) file-sharing application. The study focuses on understanding the scalability, performance, and efficiency of BitTorrent, as well as the impact of its built-in incentive mechanism on network performance. The model considers factors such as peer evolution, scalability, file-sharing efficiency, and incentives to prevent free-riding. The paper also provides numerical results based on simulations and real-world data. The fluid model captures the dynamics of BitTorrent by defining variables such as the number of downloaders (leechers) and seeds, arrival rates, and bandwidth constraints. The model is used to analyze the steady-state performance of the system, including average download time, which is found to be independent of the request arrival rate. The effectiveness of file sharing is quantified by a parameter η, which is shown to be close to 1 in BitTorrent, indicating high efficiency. The paper also explores the stability of the fluid model and its ability to capture variability in the system. A stochastic fluid model is introduced to account for fluctuations around the deterministic model. The study further examines the peer selection algorithm, which aims to encourage uploading by prioritizing peers with higher bandwidths. The analysis shows that the system's performance is influenced by the uploading bandwidth and the number of peers. The paper also discusses the impact of optimistic unchoking, a mechanism that allows peers to explore the network and avoid free-riding. It is shown that optimistic unchoking helps maintain system stability and ensures that free-riders do not gain excessive benefits. Finally, the paper presents experimental results validating the fluid model against simulations and real-world data. The results demonstrate that the fluid model accurately predicts the behavior of BitTorrent networks, even for files with a small number of completed downloads. The study concludes that BitTorrent is a scalable and efficient P2P file-sharing system, with the built-in incentive mechanisms playing a crucial role in maintaining network performance.This paper presents a fluid model and performance analysis of BitTorrent, a second-generation peer-to-peer (P2P) file-sharing application. The study focuses on understanding the scalability, performance, and efficiency of BitTorrent, as well as the impact of its built-in incentive mechanism on network performance. The model considers factors such as peer evolution, scalability, file-sharing efficiency, and incentives to prevent free-riding. The paper also provides numerical results based on simulations and real-world data. The fluid model captures the dynamics of BitTorrent by defining variables such as the number of downloaders (leechers) and seeds, arrival rates, and bandwidth constraints. The model is used to analyze the steady-state performance of the system, including average download time, which is found to be independent of the request arrival rate. The effectiveness of file sharing is quantified by a parameter η, which is shown to be close to 1 in BitTorrent, indicating high efficiency. The paper also explores the stability of the fluid model and its ability to capture variability in the system. A stochastic fluid model is introduced to account for fluctuations around the deterministic model. The study further examines the peer selection algorithm, which aims to encourage uploading by prioritizing peers with higher bandwidths. The analysis shows that the system's performance is influenced by the uploading bandwidth and the number of peers. The paper also discusses the impact of optimistic unchoking, a mechanism that allows peers to explore the network and avoid free-riding. It is shown that optimistic unchoking helps maintain system stability and ensures that free-riders do not gain excessive benefits. Finally, the paper presents experimental results validating the fluid model against simulations and real-world data. The results demonstrate that the fluid model accurately predicts the behavior of BitTorrent networks, even for files with a small number of completed downloads. The study concludes that BitTorrent is a scalable and efficient P2P file-sharing system, with the built-in incentive mechanisms playing a crucial role in maintaining network performance.