Behavior of Fast Consistency Algorithms in Large Decentralized Systems
نویسندگان
چکیده
Information replication is a major technique for improving the efficiency and autonomy of many distributed applications. Weak consistency algorithms allow us to propagate changes in an arbitrary, changing storage network in a self-organizing way. The weak consistency algorithms generate very little traffic overhead; they have low latency and are scalable, in addition to being fault tolerant. In this paper we evaluate our own weak consistency algorithm, which is called the “Fast Consistency Algorithm", and whose main aim is the propagation of changes with preference for nodes and zones of the network which have greatest demand. The algorithm has been simulated over ns-2, and evaluated with several topologies: linear, ring, random and grid, and several distributions of demand: constant, one region of high demand, two regions, and self-similar fractal mountains of demand. The impulse response of the algorithm has been characterized: the time it takes for all the network to become consistent; i.e. measuring the speed with which each node receives all events introduced at t=0 in the rest of the nodes. Employing, among other the economic concept of an utility function, we show that our “fast consistency” algorithm, interconnecting high demand zones by means of a logical topology, optimizes the distribution of changes by prioritizing the nodes with greatest demand, independently of demand distribution. In other words, it satisfies the greatest demand in the shortest amount of time.
منابع مشابه
A Framework for Performance Evaluation of Decentralized Eventual Consistency Algorithms
Eventual Consistency (EC) model is adopted by numerous large-scale distributed systems. To ensure performance and scalability, this model allows any replica to accept updates without remote synchronization. Nowadays, many EC algorithms are developed to control the behavior of the replicated data in the face of concurrent updates. Among them, those using a central server to order the updates, wh...
متن کاملAN OBSERVER-BASED INTELLIGENT DECENTRALIZED VARIABLE STRUCTURE CONTROLLER FOR NONLINEAR NON-CANONICAL NON-AFFINE LARGE SCALE SYSTEMS
In this paper, an observer based fuzzy adaptive controller (FAC) is designed fora class of large scale systems with non-canonical non-affine nonlinear subsystems. It isassumed that functions of the subsystems and the interactions among subsystems areunknown. By constructing a new class of state observer for each follower, the proposedconsensus control method solves the problem of unmeasured sta...
متن کاملDecentralized Adaptive Control of Large-Scale Non-affine Nonlinear Time-Delay Systems Using Neural Networks
In this paper, a decentralized adaptive neural controller is proposed for a class of large-scale nonlinear systems with unknown nonlinear, non-affine subsystems and unknown nonlinear time-delay interconnections. The stability of the closed loop system is guaranteed through Lyapunov-Krasovskii stability analysis. Simulation results are provided to show the effectiveness of the proposed approache...
متن کاملDecentralized Model Reference Adaptive Control of Large Scale Interconnected Systems with Time-Delays in States and Inputs
This paper investigates the problem of decentralized model reference adaptive control (MRAC) for a class of large scale systems with time varying delays in interconnected terms and state and input delays. The upper bounds of the interconnection terms are considered to be unknown. Time varying delays in the nonlinear interconnection terms are bounded and nonnegative continuous functions and thei...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2003