{"type":"conference","year":"2019","citation":{"short":"T. Götte, V.R. Vijayalakshmi, C. Scheideler, in: Proceedings of the 2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS ’19), IEEE, n.d.","ieee":"T. Götte, V. R. Vijayalakshmi, and C. Scheideler, “Always be Two Steps Ahead of Your Enemy - Maintaining a Routable Overlay under Massive Churn with an Almost Up-to-date Adversary,” in Proceedings of the 2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS ’19), Rio de Janeiro, Brazil.","apa":"Götte, T., Vijayalakshmi, V. R., & Scheideler, C. (n.d.). Always be Two Steps Ahead of Your Enemy - Maintaining a Routable Overlay under Massive Churn with an Almost Up-to-date Adversary. In Proceedings of the 2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS ’19). Rio de Janeiro, Brazil: IEEE.","ama":"Götte T, Vijayalakshmi VR, Scheideler C. Always be Two Steps Ahead of Your Enemy - Maintaining a Routable Overlay under Massive Churn with an Almost Up-to-date Adversary. In: Proceedings of the 2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS ’19). IEEE.","chicago":"Götte, Thorsten, Vipin Ravindran Vijayalakshmi, and Christian Scheideler. “Always Be Two Steps Ahead of Your Enemy - Maintaining a Routable Overlay under Massive Churn with an Almost Up-to-Date Adversary.” In Proceedings of the 2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS ’19). IEEE, n.d.","bibtex":"@inproceedings{Götte_Vijayalakshmi_Scheideler, title={Always be Two Steps Ahead of Your Enemy - Maintaining a Routable Overlay under Massive Churn with an Almost Up-to-date Adversary}, booktitle={Proceedings of the 2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS ’19)}, publisher={IEEE}, author={Götte, Thorsten and Vijayalakshmi, Vipin Ravindran and Scheideler, Christian} }","mla":"Götte, Thorsten, et al. “Always Be Two Steps Ahead of Your Enemy - Maintaining a Routable Overlay under Massive Churn with an Almost Up-to-Date Adversary.” Proceedings of the 2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS ’19), IEEE."},"conference":{"name":"2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS '19)","start_date":"20.05.19","location":"Rio de Janeiro, Brazil","end_date":"24.05.19"},"_id":"6976","file":[{"file_id":"7007","creator":"thgoette","file_size":638020,"relation":"main_file","success":1,"date_updated":"2019-01-26T16:09:08Z","content_type":"application/pdf","date_created":"2019-01-26T16:09:08Z","file_name":"Always_be_Two_Steps_Ahead_of_Your_Enemy.pdf","access_level":"closed"}],"file_date_updated":"2019-01-26T16:09:08Z","publication":"Proceedings of the 2019 IEEE 33rd International Parallel and Distributed Processing Symposium (IPDPS '19)","author":[{"last_name":"Götte","id":"34727","first_name":"Thorsten","full_name":"Götte, Thorsten"},{"last_name":"Vijayalakshmi","full_name":"Vijayalakshmi, Vipin Ravindran","first_name":"Vipin Ravindran"},{"first_name":"Christian","full_name":"Scheideler, Christian","last_name":"Scheideler","id":"20792"}],"publisher":"IEEE","date_created":"2019-01-24T18:53:11Z","has_accepted_license":"1","status":"public","abstract":[{"lang":"eng","text":"We investigate the maintenance of overlay networks under massive churn, i.e.\r\nnodes joining and leaving the network. We assume an adversary that may churn a\r\nconstant fraction $\\alpha n$ of nodes over the course of $\\mathcal{O}(\\log n)$\r\nrounds. In particular, the adversary has an almost up-to-date information of\r\nthe network topology as it can observe an only slightly outdated topology that\r\nis at least $2$ rounds old. Other than that, we only have the provably minimal\r\nrestriction that new nodes can only join the network via nodes that have taken\r\npart in the network for at least one round.\r\n Our contributions are as follows: First, we show that it is impossible to\r\nmaintain a connected topology if adversary has up-to-date information about the\r\nnodes' connections. Further, we show that our restriction concerning the join\r\nis also necessary. As our main result present an algorithm that constructs a\r\nnew overlay- completely independent of all previous overlays - every $2$\r\nrounds. Furthermore, each node sends and receives only $\\mathcal{O}(\\log^3 n)$\r\nmessages each round. As part of our solution we propose the Linearized DeBruijn\r\nSwarm (LDS), a highly churn resistant overlay, which will be maintained by the\r\nalgorithm. However, our approaches can be transferred to a variety of classical\r\nP2P Topologies where nodes are mapped into the $[0,1)$-interval."}],"user_id":"34727","ddc":["000"],"language":[{"iso":"eng"}],"date_updated":"2022-01-06T07:03:25Z","department":[{"_id":"79"}],"project":[{"_id":"1","name":"SFB 901"},{"_id":"4","name":"SFB 901 - Project Area C"},{"name":"SFB 901 - Subproject C1","_id":"13"}],"publication_status":"accepted","title":"Always be Two Steps Ahead of Your Enemy - Maintaining a Routable Overlay under Massive Churn with an Almost Up-to-date Adversary"}