Data Location In The OceanStore
Abstract:
We have examined the problem of data location in the OceanStore, a highly distributed data storage architecture. In particular, given a network of servers that contain named objects we present a probabilistic algorithm that efficiently routes requests for an object to servers believed to contain that object. This algorithm has two phases. The first phase creates a consistent distributed directory of objects using attenuated Bloom filters. In the second phase these attenuated Bloom filters are used to route object location requests.
Citations
| 5825 | Introduction to Algorithms – Cormen, Leiserson, et al. - 1992 |
| 687 | Space/time trade-offs in hash coding with allowable errors – Bloom - 1970 |
| 433 | Summary cache: A scalable wide-area web cache sharing protocol – Fan, Cao, et al. - 1998 |
| 149 | A comparison of file system workloads – Roselli, Lorch, et al. - 2000 |
| 7 | On The Maximum Degree And The Height Of A Random Recursive Tree – Szymanski - 1990 |
| 1 | A Unified Approach To Global Program Optimization – Killdal - 1973 |
| 1 | Lecture Notes on Scalar Dataflow Analysis for CS612 – Pingali - 1999 |

