Gossip-based aggregation in large dynamic networks

Abstract

As computer networks increase in size, become more heterogeneous and span greater geographic distances, applications must be designed to cope with the very large scale, poor reliability, and often, with the extreme dynamism of the underlying network. Aggregation is a key functional building block for such applications: it refers to a set of functions that provide components of a distributed system access to global information including network size, average load, average uptime, location and description of hotspots, etc. Local access to global information is often very useful, if not indispensable for building applications that are robust and adaptive. For example, in an industrial control application, some aggregate value reaching a threshold may trigger the execution of certain actions; a distributed storage system will want to know the total available free space; load balancing protocols may benefit from knowing the target average load so as to minimize the load they transfer. We propose a gossip-based protocol for computing aggregate values over network components in a fully decentralized fashion. The class of aggregate functions we can compute is very broad and includes many useful special cases such as counting, averages, sums, products and extremal values. The protocol is suitable for extremely large and highly dynamic systems due to its proactive structure-all nodes receive the aggregate value continuously, thus being able to track any changes in the system. The protocol is also extremely lightweight making it suitable for many distributed applications including peer-to-peer and grid computing systems. We demonstrate the efficiency and robustness of our gossip-based protocol both theoretically and experimentally under a variety of scenarios including node and communication failures.

P2P Evolutionary Algorithms: A Suitable Approach for Tackling Large Instances in Hard Optimization Problems

“…The challenge of tackling these large instances of a problem and the results regarding small-world structured populations in [6], drove us to analyze in this work the effects of a self-organized population using the gossiping protocol newscast [10,9]. Newscast shares some small-world properties with the WattsStrogatz model [17], such as a low average path length and a high clustering coefficient, and has been proved to scale to a large number of nodes [16].…”

Peer-to-Peer Multi-class Boosting

“…Perhaps the simplest example is gossip-based averaging [14,16], where the gossip approach is extremely robust, scalable, and efficient. However, gossip algorithms support more sophisticated algorithms that compute more complex global functions.…”

DS-Means: Distributed Data Stream Clustering

“…4. The resulting number of centroids could be different at different nodes, so a pairwise averaging algorithm [5] is run on such values. All nodes will obtain the same average number G = 1/N N i=1 G i .…”

“…To obtain this we use a simple, epidemic-based pairwise averaging algorithm [5]. After O(log N ) epidemic rounds all nodes know the average number of clusters computed by the different instances of X-means.…”