This paper presents our work in combining peer-to-peer dynamic tree management with hierarchical Operational Transformation (OT) over document trees to achieve low computational and communication costs. We discuss our approach in storing the document tree in a peer-to-peer, distributed manner and maintaining convergence, causality preservation, and intention preservation (CCI) via a peer-to-peer caching system, Because changes are sent to other users within the system only as needed (and cached when possible), our approach minimizes communication costs among multiple readers and writers. Our algorithms balance the traffic and computational load among peers. They ensure that users always have the most current/correct copy of the section(s) of the document which they are viewing. Our approach outperforms existing OT techniques that broadcast messages and compute OT for each operation at all peers. This paper presents our algorithms and simulation results demonstrating the efficiencies and load balancing among peers within the system. © Springer-Verlag Berlin Heidelberg 2007.
CITATION STYLE
Preston, J. A., & Prasad, S. K. (2007). P2P document tree management in a real-time collaborative editing system. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4873 LNCS, pp. 418–431). Springer Verlag. https://doi.org/10.1007/978-3-540-77220-0_39
Mendeley helps you to discover research relevant for your work.