, and leverages Pastry's reliability, self-organization, and locality properties. Pastry is used to create and manage groups and to build efficient multicast trees for the dissemination of messages to each group. Scribe provides best-effort reliability guarantees, but we outline how an application can extend

Abstract — The conventional wisdom has been that IP is the natural protocol layer for implementing multicast related functionality. However, more than a decade after its initial proposal, IP Multicast is still plagued with concerns pertaining to scalability, network management, deployment

directions; and 3) robust and resilient, as the partnerships enable adaptive and quick switching among multi-suppliers. We show through analysis that DONet is scalable with bounded delay. We also address a set of practical challenges for realizing DONet, and propose an efficient member- and partnership

Delay management for public transport consists of deciding whether vehicles should wait for delayed transferring passengers, with the objective of minimizing the overall passenger discomfort. We model the underlying transportation network as a directed acyclic graph, where edges represent trains

this paper. Let V be a set of train vehicles. A train v V is usually given by a train line, i.e. a sequence of stations s 1 , s 2 , . . . , s nv . The train departs at s 1 , has the intermediate stops s 2 , . . . , s nv-1 and finally arrives at s nv . Every train v is associated with a timetable t(v), a sequence t d (v, s 1 ), t a (v, s 2 ), t d (v, s 2 ), . . . , t a (v, s nv ), where t a (v, s i ) is the arrival time of train v at station s i and t d (v, s i ) is the departure time of v at station s i . Let T be the complete timetable, that is T = {t(v) : v }

on whether there are packets with more stringent delay requirements) and the granularity in delay differentiation becomes irrespective of the number and size of buffers. It is also shown that end-toend per-flow delay control can be achieved through ensuring local delay management.

The question of delay management (DM) is whether trains should wait for delayed feeder trains or should depart on time. Solutions to this problem strongly depend on the capacity constraints of the tracks making sure that no two trains can use the same piece of track at the same time. While

For many online problems disappointing lower bounds on the achiev-able competitive ratio have been shown. This statement also holds for most versions of the delay management problem in the classical online setting. In this paper we analyze how natural restrictions on the power of the adversary

The delay management problem deals with reactions in case of delays in public transportation. More specifically, the aim is to decide if connecting vehicles should wait for delayed feeder vehicles or if it is better to depart on time. As objective we consider the convenience over all customers

works must be obtained from the IEEE. SERVER-SIDE PLAYOUT DELAY MANAGEMENT FOR VIDEO STREAMING