An adaptive data placement scheme for scaleable object storage system

Yihui Luo; Wei Deng

doi:10.1117/12.823263

19 March 2009 An adaptive data placement scheme for scaleable object storage system

Yihui Luo, Wei Deng

Proceedings Volume 7125, Eighth International Symposium on Optical Storage and 2008 International Workshop on Information Data Storage; 712527 (2009) https://doi.org/10.1117/12.823263
Event: Eighth International Symposium on Optical Storage and 2008 International Workshop on Information Data Storage, 2008, Wuhan, China

Abstract

Scaleable storage system has contradiction between usability and extendibility, so a suitable data placement scheme is given to solve this problem in scaleable object storage system. This scheme includes data placement algorithm and data reorganization algorithm. The data placement scheme has the following characteristics: the data placement algorithm uses key as the seed of random data generator to map data objects to storage nodes, which can provide reliable accessing to scaleable storage system. The data placement algorithm distributes data objects uniformly among storage nodes, which can improve I/O performance by accessing simultaneously several nodes. The data reorganization algorithm migrates data only from old storage nodes to added storage nodes (or from withdrawn nodes to surplus nodes), which can improve usability by reducing data migration. At last, the two algorithms have less cost of time and space, which is suitable to large-scale storage systems.

Citation Download Citation

Yihui Luo and Wei Deng "An adaptive data placement scheme for scaleable object storage system", Proc. SPIE 7125, Eighth International Symposium on Optical Storage and 2008 International Workshop on Information Data Storage, 712527 (19 March 2009); https://doi.org/10.1117/12.823263

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available