Multicomputer object stores: the Distributed Texas experiment
Download:
|
|
by Stephen M Blackburn, Robin B Stanton
ftp://ftp.cis.upenn.edu/pub/nettles/pos/letter/US21-stanton.ps
Add To MetaCart
Abstract:
The paper is about design of object stores for distributed memory computers. Problems of distribution within such stores are described through the design of Distributed Texas, a system derived from the Texas store [6] and implemented on a scalable multicomputer (a Fujitsu AP1000). Problems discussed include creating a single image of an object space across multicomputer architectures, tradeoffs between page-based and object-based granularity, the distribution of store level services and the management of concurrency and persistance within those distributed services. Problems of building layered software on multicomputers pervade these issues. Preliminary scalability performance figures for Distributed Texas are also presented.
Citations
| 248 | The OO7 benchmark – Carey, DeWitt, et al. - 1993 |
| 167 | Shoring Up Persistent Applications – Carey, DeWitt, et al. - 1994 |
| 109 | Texas: An Efficient, Portable Persistent Store – Singhal, Kaddak, et al. - 1992 |
| 77 | Pointer Swizzling at Page Fault Time: Efficiently Supporting Huge Address Spaces on Standard Hardware – Wilson - 1991 |
| 55 | The architecture of the EXODUS extensible DBMS – CAREY, DEWITT, et al. - 1986 |
| 20 | CAP-II Architecture – Ishihata, Horie, et al. - 1990 |
| 20 | The HiDIOS Filesystem – Tridgell, Walsh - 1995 |
| 18 | Orthogonally persistent systems – Atkinson, Morrison - 1995 |
| 3 | Implementing MPI using interrupts and remote copying for the AP1000/AP1000 – Sitsky - 1995 |
