A Cost-Benefit Approach to Fault Tolerant Communication and Information Access

8/14/01


Click here to start


Table of Contents

A Cost-Benefit Approach to Fault Tolerant Communication and Information Access

Project Objective

Network Fault Models: Global Network Stability

Network Fault Models: Partial Network Stability

Network Faults Models: Totally Unstable network

Project Failures Models (requested by Doug)

Outline

Unstable Network Routing

The Gravitational Flow Approach

Example

Example (cont.)

Gravitational Flow Routing

Outline

The Cost-Benefit Philosophy

The Cost-Benefit Framework for Heterogeneous Resources.

Cost-Benefit based Flow Control

Global Flow Control for Overlay Networks

Global Flow Control for Overlay Networks (cont.)

Outline

Spread Overlay Network over CAIRN

A Cost-Benefit Approach to Information Access in a Cluster

Load Balancing A Web Cluster

mod_backhand in Action

Wackamole

Wackamole

Wackamole

Wackamole: Dynamic Cluster with N-way Fail-over

Replication

An Architecture for Active (Peer) Database Replication

An Architecture for Active (Peer) Database Replication (cont.)

A Generic Replication Engine

The Postgres Interceptor

Deliverables in the last 6 months

Author: Yair Amir & Baruch Awerbuch

Email: yairamir@cs.jhu.edu

Home Page: http://www.cnds.jhu.edu/funding/tolerant_networks/

Other information:
A Cost-Benefit Approach to Fault Tolerant Communication and Informantion Access