Massively parallel I/O: Building an infrastructure for parallel computing
The solution of Grand Challenge Problems will require computations that are too large to fit in the memories of even the largest machines. Inevitably, new designs of I/O systems will be necessary to support them. This report describes the work in investigating I/O subsystems for massively parallel computers. Specifically, the authors investigated out-of-core algorithms for common scientific calculations present several theoretical results. They also describe several approaches to parallel I/O, including partitioned secondary storage and choreographed I/O, and the implications of each to massively parallel computing.
- Research Organization:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE Office of Energy Research, Washington, DC (United States)
- DOE Contract Number:
- AC04-94AL85000
- OSTI ID:
- 468615
- Report Number(s):
- SAND-96-2672; ON: DE97004908; TRN: AHC29710%%111
- Resource Relation:
- Other Information: PBD: Apr 1997
- Country of Publication:
- United States
- Language:
- English
Similar Records
Beyond core: Making parallel computer I/O practical
Beyond core: Making parallel computer I/O practical
High speed networks, visualization, and massive parallelism in the Advanced Computing Laboratory
Conference
·
Sat May 01 00:00:00 EDT 1993
·
OSTI ID:468615
+1 more
Beyond core: Making parallel computer I/O practical
Conference
·
Fri Jan 01 00:00:00 EST 1993
·
OSTI ID:468615
+1 more
High speed networks, visualization, and massive parallelism in the Advanced Computing Laboratory
Conference
·
Sat Aug 01 00:00:00 EDT 1992
·
OSTI ID:468615
+2 more