Beyond striping; The Bridge multiprocessor file system
- Rochester Univ., NY (USA). Dept. of Computer Science
High-performance parallel computers require high-performance file systems. Exotic I/O hardware will be of little use if file system software runs on a single processor of a many-processor machine. The authors believe that cost-effective I/O for large multiprocessors can best be obtained by spreading both data and file system computation over a large number of processors and disks. To assess the effectiveness of this approach, they have implemented a prototype system called Bridge, and have studied its performance on several data intensive applications, among them external sorting. A detailed analysis of their sorting algorithm is presented. It indicates that Bridge can profitably be used on configurations in excess of one hundred processors with disks. Empirical results on a 32-processor implementation agree with the analysis, providing a high degree of confidence in this prediction. It is argued that file systems such as Bridge will satisfy the I/O needs of a wide range of parallel architectures and applications.
- OSTI ID:
- 6996242
- Journal Information:
- Computer Architecture News; (USA), Vol. 17:5; ISSN 0163-5964
- Country of Publication:
- United States
- Language:
- English
Similar Records
The Performance of Parallel Disk Write Methods for Linux Multiprocessor Nodes
Center for Technology for Advanced Scientific Componet Software (TASCS)