Networking and Communication Engineering

K . Ramesh Kumar ¹, R. S. D. Wahida Banu ²

Affiliations
1 Bharat Sanchar Nigam Limited, Tamilnadu, IN
2 Government College of Engineering, Salem, IN

Abstract

As regeneration capability is not available in all optical networks, the physical layer impairments (PLI) accumulate in the network. During transmission, Quality of Service (QoS) of signals as measured by their Bit Error Rates (BER) is degraded due to propagation in the fibers, amplifier noise, and crosstalk between channels arising from the switches and demultiplexers which cannot be removed at the physical layer. It is possible to mitigate physical layer effects at the network layer using appropriate QoS aware Wavelength Assignment (WA) algorithm. In this paper a WA Scheme based on wavelength pre ordering is presented with dedicated path protection consideration that mitigates PLI. The wavelength preordering technique expected to improve BER performances, due to crosstalk reductions. The proposed WA scheme is performed on a standard regional area network with standard network parameters by defining a realistic physical layer transmission model. The proposed technique is evaluated using simulations and simulations show that the proposed algorithm performs better in terms of blocking rate and BER.

Keywords

Routing and Wavelength Assignment (RWA), Physical Layer Impairments, Optical Networks, Dedicated Path Protection, Cross Talk, QoS Aware, Wavelength Assignment (WA).

Full Text

     

M. Nithya ¹, R. S. D. Wahida Banu ²

Affiliations
1 Anna University, Chennai, IN
2 Govt Coll of Engg, Salem, IN

Abstract

Grid technologies enable large-scale sharing of resources within formal or informal consortia of individuals and/or institutions: what are sometimes called virtual organizations. In these settings, the discovery, characterization, and monitoring of resources, services, and computations can be challenging due to the considerable diversity ,large numbers, dynamic behavior, and geographical distribution of the entities in which a user might be interested. Authorization and access control; the significant aspects of security, have attracted increased attention in grid computing. In authorization an essential part to protect the information and all other resources from unauthorized access by controlling the access via a particular system. Role-Based Access Control (RBAC) is a fairly novel access control technique that provides a centralized, dynamic, and consistent authorization management specifically for the requirements of a particular industry to improve its security. RBAC was developed in the Globus toolkit(GT). Access control is formed by the support of Community Authorization Service (CAS) and this CAS was used and developed in many problems grid research. The major problem with the CAS is that the user credentials are revealed to the virtual organization (VO).In this paper, analyzes the unique security requirements grid computing and develops a security policy and a corresponding security architecture. We also evaluate the performance of Role Based access control architecture.

Keywords

Grid Computing, Grid Security, Authorization and Access Control, Role Based Access Control (Rbac), Globus Toolkit (Gt), Community Authorization Server (Cas), Virtual Organization (Vo)S.

Full Text

     

M. Azath ¹, R. S. D. Wahida banu ¹

Affiliations
1 Anna University, Coimbatore, IN

Abstract

In this paper we propose a bursty traffic splitting algorithm splits the traffic flows over multiple parallel paths, based on a split vector. In this algorithm, instead of switching packets or flows, it switches packet bursts. Since the packet bursts are smaller in size, the algorithm splits the traffic dynamically and accurately. At the same time, the condition forced on their latency difference, ensures that no packets are reordered. To achieve fair bandwidth allocations, load balancing is attained in the system since the high-rate aggressive traffic flows are splitted along multiple parallel paths. The proposed switching technique is executed in the edge and core routers. We will show by simulations that our adaptive packet scheduler performs better than the standard fair-queuing techniques.

Keywords

Load Balancing, Splitting, Bursty Traffic, Aggressive Flows, Scheduler, RED.

Full Text

     

A. Neela Madheswari ¹, R. S. D. Wahida Banu ¹

Affiliations
1 Anna University, Coimbatore, IN

Abstract

Cluster computing is best characterized as the integration of a number of off-the-shelf commodity computers and resources integrated through hardware, networks, and software to behave as a single computer. The processes that are running on different processors on clusters may be in need of communicated with each other. If the processes that are going to communicate are not allotted to a single processor may suffer some performance penalties due to the communication overhead. A new scheduling algorithm is proposed using proper job grouping which is mainly based on the communication cost between the processes and are allotted to the same processor in the clusters and thus the performance is tuned such that the time taken for communication overhead is avoided. The experimental setup is done using Java and the results are analyzed which proves the effectiveness of the algorithm.

Keywords

Scheduling, Bandwidth, Cache Memory, Scheduling Algorithms.

Full Text