Abstract
We study a scenario in which n nodes of a mobile ad-hoc network continuously collect data. Their task is to repeatedly update aggregated information about the data, e.g., the maximum, the sum, or the full information about all data received by all nodes at a given time step. This aggregated information has to be disseminated to all nodes.
We propose two performance measures for distributed algorithms for these tasks: The delay is the maximum time needed until the aggregated information about the data measured at some time is output at all nodes. We assume that a node can broadcast information proportional to a constant number of data items per round. A too large communication volume needed for producing an output can lead to the effect that the delay grows unboundedly over time. Thus, we have to cope with the restriction that outputs are computed not for all but only for a fraction of rounds. We refer to this fraction as the output rate of the algorithm.
Our main technical contributions are trade-offs between delay and output rate for aggregation problems under the assumption of T-stable dynamics in the mobile ad-hoc network: The network is always connected and is stable for time intervals of length 
where 
is the time needed to compute a maximal independent set. For the maximum function, we are able to show that we can achieve an output rate of 
with delay 
. For the sum, we show that it is possible to achieve an output rate of 
with delay 
if 
, and if 
, we can achieve an output rate of 
with delay 
.
This work was partially supported by the German Research Foundation (DFG) within the Priority Program “Algorithms for Big Data” (SPP 1736), by the EU within FET project MULTIPLEX under contract no. 317532, and the International Graduate School “Dynamic Intelligent Systems”.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Abshoff, S., Benter, M., Cord-Landwehr, A., Malatyali, M., Meyer auf der Heide, F.: Token dissemination in geometric dynamic networks. In: Flocchini, P., Gao, J., Kranakis, E., Meyer auf der Heide, F. (eds.) ALGOSENSORS 2013. LNCS, vol. 8243, pp. 22–34. Springer, Heidelberg (2014)
Abshoff, S., Benter, M., Malatyali, M., Meyer auf der Heide, F.: On two-party communication through dynamic networks. In: Baldoni, R., Nisse, N., van Steen, M. (eds.) OPODIS 2013. LNCS, vol. 8304, pp. 11–22. Springer, Heidelberg (2013)
Awerbuch, B.: Optimal distributed algorithms for minimum weight spanning tree, counting, leader election and related problems (detailed summary). In: Aho, A.V. (ed.) STOC, pp. 230–240. ACM (1987)
Chaudhuri, S., Dubhashi, D.P.: Probabilistic recurrence relations revisited. Theor. Comput. Sci. 181(1), 45–56 (1997)
Cornejo, A., Gilbert, S., Newport, C.C.: Aggregation in dynamic networks. In: Kowalski, D., Panconesi, A. (eds.) PODC, pp. 195–204. ACM (2012)
Dutta, C., Pandurangan, G., Rajaraman, R., Sun, Z., Viola, E.: On the complexity of information spreading in dynamic networks. In: Khanna, S. (ed.) SODA, pp. 717–736. SIAM (2013)
Haeupler, B., Karger, D.R.: Faster information dissemination in dynamic networks via network coding. In: Gavoille, C., Fraigniaud, P. (eds.) PODC, pp. 381–390. ACM (2011)
Haeupler, B., Kuhn, F.: Lower bounds on information dissemination in dynamic networks. In: Aguilera, M.K. (ed.) DISC 2012. LNCS, vol. 7611, pp. 166–180. Springer, Heidelberg (2012)
Karp, R.M.: Probabilistic recurrence relations. J. ACM 41(6), 1136–1150 (1994)
Kuhn, F., Locher, T., Schmid, S.: Distributed computation of the mode. In: Bazzi, R.A., Patt-Shamir, B. (eds.) PODC, pp. 15–24. ACM (2008)
Kuhn, F., Locher, T., Wattenhofer, R.: Tight bounds for distributed selection. In: Gibbons, P.B., Scheideler, C. (eds.) SPAA, pp. 145–153. ACM (2007)
Kuhn, F., Lynch, N.A., Oshman, R.: Distributed computation in dynamic networks. In: Schulman, L.J. (ed.) STOC, pp. 513–522. ACM (2010)
Linial, N.: Locality in distributed graph algorithms. SIAM J. Comput. 21(1), 193–201 (1992)
Luby, M.: A simple parallel algorithm for the maximal independent set problem. SIAM J. Comput. 15(4), 1036–1053 (1986)
Mosk-Aoyama, D., Shah, D.: Computing separable functions via gossip. In: Ruppert, E., Malkhi, D. (eds.) PODC, pp. 113–122. ACM (2006)
Panconesi, A., Srinivasan, A.: Improved distributed algorithms for coloring and network decomposition problems. In: Kosaraju, S.R., Fellows, M., Wigderson, A., Ellis, J.A. (eds.) STOC, pp. 581–592. ACM (1992)
Schneider, J., Wattenhofer, R.: A log-star distributed maximal independent set algorithm for growth-bounded graphs. In: Bazzi, R.A., Patt-Shamir, B. (eds.) PODC, pp. 35–44. ACM (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Abshoff, S., Meyer auf der Heide, F. (2014). Continuous Aggregation in Dynamic Ad-Hoc Networks. In: Halldórsson, M.M. (eds) Structural Information and Communication Complexity. SIROCCO 2014. Lecture Notes in Computer Science, vol 8576. Springer, Cham. https://doi.org/10.1007/978-3-319-09620-9_16
Download citation
DOI: https://doi.org/10.1007/978-3-319-09620-9_16
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-09619-3
Online ISBN: 978-3-319-09620-9
eBook Packages: Computer ScienceComputer Science (R0)