Research and design of deployment framework for blade-based data center

Authors:
Haiping Qu;Xiuwen Wang;Lu Xu;Jiangang Zhang;Xiaoming Han
Affiliations:
,Institute of Computer Technology, Chinese Academy of Sciences, Beijing, China;,Institute of Computer Technology, Chinese Academy of Sciences, Beijing, China;Institute of Computer Technology, Chinese Academy of Sciences, Beijing, China;Institute of Computer Technology, Chinese Academy of Sciences, Beijing, China;Institute of Computer Technology, Chinese Academy of Sciences, Beijing, China
Venue:
ICA3PP'10 Proceedings of the 10th international conference on Algorithms and Architectures for Parallel Processing - Volume Part I
Year:
2010

Citing 3
Cited 0

Dynamic Virtual Clusters in a Grid Site Manager

HPDC '03 Proceedings of the 12th IEEE International Symposium on High Performance Distributed Computing
NVD: the Network Virtual Device for HA-SonD

IWNAS '06 Proceedings of the 2006 International Workshop on Networking, Architecture, and Storages
VSVM-enhanced: a Volume Manager Based on the EVMS Framework

GCCW '06 Proceedings of the Fifth International Conference on Grid and Cooperative Computing Workshops

Quantified Score

Hi-index	0.00

Visualization

Abstract

For those outsourcing data centers hosting different applications or services, blade server is becoming the most popular solution But the deployment of systems in such data centers will be a time-consuming and cumbersome task Moreover, the service provided by the servers in data center changed frequently, which require a flexible service management system to reduce the burden of the system administrator This paper presents the Bladmin (Blade Admin) system to provide dynamic, flexible and quick service and operation system deployment It can build a Virtual Environment (VE) for each service All the nodes in the VE serves for one certain kind of application and some nodes have to be shift from one VE to another to adopt the varying workloads Experiment results demonstrate that system-level performance has been greatly improved and the running of one Paradigm job also show the efficiency of Bladmin framework.