Algorithms for E cient Filtering in Content Based Multicast

Content Based Multicast is a type of multicast where the source sends a set of di erent classes of information and not all the sub scribers in the multicast group need all the information Use of ltering publish subscribe agents on the intermediate nodes was suggested to lter out the unnecessary information on the multicast tree However lters have their own drawbacks like processing delays and infrastruc ture cost Hence it is desired to place these lters most e ciently An O n dynamic programming algorithm was proposed to calculate the best locations for lters that would minimize overall delays in the net work We propose an improvement of this algorithm which exploits the geometry of piecewise linear functions and fast merging of sorted lists represented by height balanced search trees to achieve O n log n time complexity Also we show an improvement of this algorithm which runs in O n log h time where h is the height of the multicast tree This problem is closely related to p median and uncapacitated facility loca tion over trees Theoretically this is an uncapacitated analogue of the p inmedian problem on trees as de ned in