Dynamic Multicast in Overlay Networks with Linear Capacity Constraints

Abstract

Abstract-In a peer-to-peer overlay network, the phenomonon of multiple overlay links sharing bottleneck physical links leads to correlation of overlay link capacities. We are able to more accurately model the overlay by incorporating these linear capacity constraints (LCC). We formulate the problem of maximizing bandwidth in overlay multicast using our LCC model. We show that finding a maximum-bandwidth multicast tree in an overlay network with LCC is NP-complete. Therefore, an efficient heuristics algorithm is designed to solve the problem. Extensive simulations show that our algorithm is able to construct multicast trees that are optimal or extremely close to optimal, with significantly higher bandwidth than trees formed in overlays with no LCC. Furthermore, we develop a fully distributed algorithm for obtaining near-optimal multicast trees, by means of gossip-based algorithms and a restricted but inherently distributed class of LCC (node-based LCC). We demonstrate that the distributed algorithm converges quickly to the centralized optimal and is highly scalable.

The Design and Application of an Improved Word Segmentation Algorithm Based on PATRICIA Tree Dictionary in Communication Material

“…The new mechanism is composed of two parts, as shown in Figure .1: Hash table (1) the first word of the hash table GB Chinese characters based on area code, simply step hashed can find Chinese characters in the first word of the hash table entry in the serial number (2) the number of 4-byte word: the word the number of the first word (3) the first entry in the pointer 4 bytes: point to the root of the word Chinese characters corresponding PATRICIA tree in communication material.…”

“…Prefix entry described below to improve query. Prefix words, part of the bit string is the same, so the path in PATIACIA tree partially overlap each keyword deposited PATIACIA end with '/ 0' path '/ 0 A 'disposal fork which compares bit 16 * n +1,16 is a bit of a character string number, n is the number of words, comparative 16 * n +1 internal node NODE [1], entry in the leftmost leaf node of node [2] ---NODE [n] is the prefix entries in all the leaf nodes of the node on the right sub-tree, because the comparison bit is incremented in each sub-tree, the search process on a "first root order, for example, in the figure above, the comparison bit 17 nodes in the left-most leaf node in the" big "is the prefix of the node leaf node of the right sub tree another example, more for "33" node, you can get the two prefix "major" and "white", they are "major cases" and "cabbage" prefix. Increase on this basis a major and how three words, only four terms in the major and returns a 4, then the major and a prefix word, while the presence of a four word phrase, To read two words, four words after the major match.…”

“…But should also be soberly aware that, although these methods has its own characteristics, in some ways demonstrated its unique performance, but these methods always there is still lack of one kind or another [2] . If these different methods to combine each other, complement each other, and no doubt will certainly be able to improve the performance of the system.…”

Balanced Link Mapping in Multicast Service Overlay Networks Design

“…The authors argue that for overlay proxy placement, the greedy algorithm is a practical and efficient solution, and for overlay link selection, we need to consider the trade-off between end user delay and tree cost. In [5] an efficient heuristics algorithm is designed to solve the problem of maximizing bandwidth in overlay multicast using linear capacity constraints model. However, all above researches mainly optimize construction cost and end-to-end delay, or consider the trade-off between tree cost and end user delay.…”