On the modeling and performance characteristics of a serpentine tape drive
Proceedings of the 1996 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Component software: beyond object-oriented programming
Component software: beyond object-oriented programming
Proceedings of the 2001 ACM/IEEE conference on Supercomputing
Access Time Estimation for Tertiary Storage Systems
Euro-Par '02 Proceedings of the 8th International Euro-Par Conference on Parallel Processing
Application of Component-Expert Technology for Selection of Data-Handlers in CrossGrid
Proceedings of the 9th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Data Grids, Collections, and Grid Bricks
MSS '03 Proceedings of the 20 th IEEE/11 th NASA Goddard Conference on Mass Storage Systems and Technologies (MSS'03)
Future Generation Computer Systems
An efficient and transparent transaction management based on the data workflow of HVEM DataGrid
CLADE '08 Proceedings of the 6th international workshop on Challenges of large applications in distributed environments
Grid Enabled Virtual Storage System Using Service Oriented Architecture
Proceedings of the 2005 conference on Software Engineering: Evolution and Emerging Technologies
Future Generation Computer Systems
GCCB'06 Proceedings of the 2006 international conference on Distributed, high-performance and grid computing in computational biology
Analysis of interoperability issues between EGEE and VEGA grid infrastructures
HPCC'06 Proceedings of the Second international conference on High Performance Computing and Communications
Implementation of replication methods in the grid environment
EGC'05 Proceedings of the 2005 European conference on Advances in Grid Computing
Stochastic approach for secondary storage data access cost estimation
EGC'05 Proceedings of the 2005 European conference on Advances in Grid Computing
| Hi-index | 0.00 |
In typical Data Grids large amounts of replicated dataare stored all over the globe in different storage systemswith access latencies ranging from seconds to hours. Thetask of a replica management system is not only to keeptrack of the replicas but also to select those replicas thatcan be accessed by an application program with a minimalresponse or transfer time. Most wide-area replicationresearch focuses on network-based replica selection. However,our past experience with Data Grids has shown thatoften hierarchical storage systems are the main bottleneckrather than network links. This is due to the fact that accesslatencies of hierarchical storage systems can be of theorder of seconds up to hours in case the data resides on atape that is not mounted yet.In this paper we give an overview of our replica managementframework called Reptor and a storage system costestimator that is used. Furthermore, we give details on accessestimation of file replicas that reside on hierarchicalstorage systems. The results show that the access estimatesprovide a good basis for a replica management system toperform efficient replica selection.