Abstract
Conventional migration strategies attempt to evenly balance the load across all available server machines. This paper discusses why conventional migration approaches are not necessarily appropriate for distributed memory-based file systems and presents and alternative approach that spreads data (possibly unevenly) across as few machines as possible and involves other available machines only as needed. The main advantage of our approach is that it keeps the system minimally distributed thereby reducing the failure rate among servers, the communication overhead among servers, the time needed to compute data relocation, distributed addressing costs, and the probability of unanticipated migrations (e.g., caused by, and an inconvenience to, returning users).
Original language | English |
---|---|
Pages | 151-160 |
Number of pages | 10 |
State | Published - 1997 |
Event | Proceedings of the 1997 7th International Workshop on Research Issues in Data Engineering, RIDE'97 - Birmingham, UK Duration: Apr 7 1997 → Apr 8 1997 |
Conference
Conference | Proceedings of the 1997 7th International Workshop on Research Issues in Data Engineering, RIDE'97 |
---|---|
City | Birmingham, UK |
Period | 4/7/97 → 4/8/97 |
ASJC Scopus subject areas
- Software
- Engineering (miscellaneous)
- Hardware and Architecture