Research Article
Topology Discovery and Performance Information Services for Optical Grids
@INPROCEEDINGS{10.1109/BROADNETS.2007.4550415, author={L. Valcarenghi and F. Paolucci and P. Castoldi and F. Cugini and D. Adami and D. Ficara and S. Giordano}, title={Topology Discovery and Performance Information Services for Optical Grids}, proceedings={4th International IEEE Conference on Broadband Communications, Networks, Systems}, proceedings_a={BROADNETS}, year={2008}, month={6}, keywords={Bandwidth Computer networks Grid computing IP networks Monitoring Network topology Optical computing Telecommunication traffic Tomography Web and internet services}, doi={10.1109/BROADNETS.2007.4550415} }- L. Valcarenghi
F. Paolucci
P. Castoldi
F. Cugini
D. Adami
D. Ficara
S. Giordano
Year: 2008
Topology Discovery and Performance Information Services for Optical Grids
BROADNETS
IEEE
DOI: 10.1109/BROADNETS.2007.4550415
Abstract
Global Grid Computing goal is to connect heterogeneous computational resources belonging to the same Virtual Organization (VO) through the Internet to form a single, more powerful virtual computer. However, to fulfill this goal it is necessary to develop services that provide the virtual computer with the same functionalities of individual end systems, such as security, interprocess communication, and resource management. This paper deals with a specific subset of the aforementioned services: topology discovery and performance information services. These services provide the Grid clients (i.e., the end users or the applications) with information about the topology of the network infrastructure connecting the end-systems belonging to a VO and some details on the network performance, such as link bandwidth capacity and delay. First, an overview of the current tools for monitoring network infrastructures supporting grid-enabled applications is presented. Then, new advanced tools for collecting information about the network topology supporting grid-enabled applications are proposed. For each tool the issues to be dealt with and its architecture and implementation are outlined. Finally their performance are experimentally evaluated.
