A deadlock-free dynamic reconfiguration protocol for distributed routing on interconnection networks

نویسنده

  • Mohiadeen Abdul Kadhar
چکیده

In interconnection networks, reconfiguration protocol is necessary to remap and reconnect the network paths, so that the network remains connected. However, the reconfiguration process brings the deadlock problem and prevention of deadlock is a tedious task in this situation. In existing works, very little work have considered deadlock problem and further, they paid no attention to reduce packet loss rate. In this paper, we propose a Token-Based (TB) robust deadlock-free dynamic reconfiguration protocol. When a device observes topology changes or detects faulty nodes, it triggers the reconfiguration process and it becomes the Reconfiguration Controller (RC). Initially, HELLO message is transmitted by the RC to all devices for which they respond with a network status message. The RC constructs the new routing function based on the received network status messages. To synchronize the old and new routing functions, the RC distributes Reconfiguration Token (RT) in an ordered way. First, it distributes to the devices that surrounds the failed device and then to other devices. Every device holds the packet until it gets packet according to new routing function and then starts the transmission. By simulation, we show efficacy of our reconfiguration protocol. The network evaluation parameter like throughput, latency (delay) and pocket loss are measured in the high and low load scenarios in NS2 network simulator. We compare the results with existing protocol Overlapping Static Reconfiguration (OSR). Based on the simulation results we have proved that the proposed token based reconfiguration protocol is produces better efficiency in all aspects.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Modularity, Routing and Fault Tolerance in Interconnection Networks

In interconnection networks, there are several problems related to deadlock. We shall address deadlocks in the context of modularity, shortest path routing in irregular networks and fault tolerance. Deadlock freedom in interconnection networks is not a modular issue. One cannot generally combine two or more deadlock free interconnection network configurations without introducing new deadlock pr...

متن کامل

Part I: A Theory for Deadlock-Free Dynamic Network Reconfiguration

This paper develops theoretical support useful for determining deadlock properties of dynamic network reconfiguration techniques and also serves as a basis for the development of design methodologies useful for deriving deadlock-free reconfiguration techniques. It is applicable to interconnection networks typically used in multiprocessor servers, network-based computing clusters and distributed...

متن کامل

Simple Deadlock-Free Dynamic Network Reconfiguration

Dynamic reconfiguration of interconnection networks is defined as the process of changing from one routing function to another while the network remains up and running. The main challenge is in avoiding deadlock anomalies while keeping restrictions on packet injection and forwarding minimal. Current approaches fall in one of two categories. Either they require the existence of extra network res...

متن کامل

Up-Down Routing Based Deadlock Free Dynamic Reconfiguration in High Speed Local Area Networks

Dynamic reconfiguration of high speed switched network is the process of changing from one routing function to another while the network remains in running mode. Current distributed switch-based interconnected systems require high performance, reliability and availability. These systems changes their topologies due to hot expansion of components, link or node activation and deactivation. Theref...

متن کامل

A Protocol for Deadlock-Free Dynamic Reconfiguration in High-Speed Local Area Networks

ÐHigh-speed local area networks (LANs) consist of a set of switches interconnected by point-to-point links, and hosts linked to those switches through a network interface card. High-speed LANs may change their topology due to switches being turned on/off, hot expansion, link remapping, and component failures. In these cases, a distributed reconfiguration protocol analyzes the topology, computes...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • Int. Arab J. Inf. Technol.

دوره 11  شماره 

صفحات  -

تاریخ انتشار 2014