Title :
Lake: Towards Highly Manageable Cluster Storage for Extremely Scalable Services
Author :
Park, Keun-Tae ; Kim, Hong-Yeon ; Kim, Young-Chul ; Lee, Sang-Min ; Kim, Young-Kyun ; Kim, Myung-Joon
Author_Institution :
Dept. of Internet Platform Res., Electron. & Telecommun. Res. Inst., Seoul
fDate :
June 30 2008-July 3 2008
Abstract :
Lake is a storage system that efficiently clusters storage servers in a manner that both scalability and availability are enhanced. Ultimately, it targets on an Exa-bytes scale cluster storage. As a practical cluster file system, Lake file system (LakeFS) not only focuses on the scalable performance enhancement with clustering but also provides manageability essential to bind and control up to thousands of storage servers. We implemented the LakeFS on the Linux OS and measured the performance on the small cluster system.The result shows notable performance enhancement by LakeFS.
Keywords :
Linux; software reliability; storage management; Lake file system; Linux OS; availability; exa-bytes scale cluster storage; scalability; scalable services; storage system; Availability; Bandwidth; Control systems; File servers; File systems; Lakes; Network servers; Scalability; Web and internet services; Web server; Cluster File System; Manageability; Scalability;
Conference_Titel :
Computational Sciences and Its Applications, 2008. ICCSA '08. International Conference on
Conference_Location :
Perugia
Print_ISBN :
978-0-7695-3243-1
DOI :
10.1109/ICCSA.2008.37
