Results 1 
2 of
2
Gossip versus Deterministic Flooding: Low Message Overhead and High Reliability for Broadcasting on Small Networks
"... Rumor mongering (also known as gossip) is an epidemiological protocol that implements broadcasting with a reliability that can be very high. Rumor mongering is attractive because it is generic, scalable, adapts well to failures and recoveries, and has a reliability that gracefully degrades with t ..."
Abstract

Cited by 41 (0 self)
 Add to MetaCart
Rumor mongering (also known as gossip) is an epidemiological protocol that implements broadcasting with a reliability that can be very high. Rumor mongering is attractive because it is generic, scalable, adapts well to failures and recoveries, and has a reliability that gracefully degrades with the number of failures in a run. However, rumor mongering uses random selection for communications. We study the impact of using random selection in this paper. We present a protocol that superficially resembles rumor mongering but is deterministic. We show that this new protocol has most of the same attractions as rumor mongering. The one attraction that rumor mongering hasnamely graceful degradationcomes at a high cost in terms of the number of messages sent. We compare the two approaches both at an abstract level and in terms of how they perform in an Ethernet and small wide area network of Ethernets.
OrderP: An Algorithm to Order Network Partitionings
"... this paper, we propose an algorithm that orders network partitionings in decreasing order of probability. This algorithm is similar to the Most Probable State Enumeration (MPSE) algorithm proposed by Li and Silvester. By looking at only the most probable partitionings, a good estimate of the perform ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
this paper, we propose an algorithm that orders network partitionings in decreasing order of probability. This algorithm is similar to the Most Probable State Enumeration (MPSE) algorithm proposed by Li and Silvester. By looking at only the most probable partitionings, a good estimate of the performance of the network may be obtained. This approach also gives bounds on the network performance. Two distinct goals, both equally important, have been attained in this paper. First we have proposed an algorithm OrderP, to generate network partitionings in the order of their decreasing probabilities of occurrence. We are not aware of any other algorithm that specifically orders network partitionings. However, three algorithms (Order, OrderII, and NewOrder) that generate network states in order of their decreasing state probabilities are known, for dualmode (up/down) components. We improvised these algorithms to generate the most probable network partitionings and then made a comparison with algorithm OrderP. OrderP has a lower computational complexity than all of the three improvised algorithms. The second goal has been to apply the algorithm in the real world and demonstrate its value in performance modeling of distributed systems. We believe that with the development of this algorithm, performance modeling of fairly large systems, hitherto unattempted due to large computational costs will be feasible. Although the obtained probabilities of occurrence of various partitionings using OrderP, are estimates of their real values, we claim that the approximation is close enough and substantiate this claim with an example. A methodology to compute performance measures under conditions of network partitioning has also been proposed, and corroborated with examples in the domain ...