Graduate School and Research Center in Digital Sciences

On scheduling and redundancy for P2P backup

Toka, Laszlo; Dell'Amico, Matteo; Michiardi, Pietro

CoRR 2010, ArXiv:1009.1344v1

                        An online backup system should be quick and       reliable in both saving and restoring users' data. To do so in       a peer-to-peer implementation, data transfer scheduling and the       amount of redundancy must be chosen wisely. We formalize the       problem of exchanging multiple pieces of data with intermittently       available peers, and we show that random scheduling completes       transfers nearly optimally in terms of duration as long as the       system is sufficiently large. Moreover, we propose an adaptive       redundancy scheme that improves performance and decreases       resource usage while keeping the risks of data loss low. Extensive       simulations show that our techniques are effective in a realistic       trace-driven scenario with heterogeneous bandwidth.

Document Arxiv Bibtex

Title:On scheduling and redundancy for P2P backup
Department:Data Science
Eurecom ref:3238
Copyright: DBLP
Bibtex: @inproceedings{EURECOM+3238, year = {2010}, title = {{O}n scheduling and redundancy for {P}2{P} backup}, author = {{T}oka, {L}aszlo and {D}ell'{A}mico, {M}atteo and {M}ichiardi, {P}ietro}, booktitle = {{C}o{RR} 2010, {A}r{X}iv:1009.1344v1}, address = {}, month = {09}, url = {} }
See also: