A framework for scheduling parallel dbms user-defined programs on an attached high-performance computer

Authors:
Michael A. Kochte;Ramesh Natarajan
Affiliations:
Universitaet Stuttgart, Stuttgart, Germany;IBM Thomas J. Watson Research Center, Yorktown Heights, NY, USA
Venue:
Proceedings of the 5th conference on Computing frontiers
Year:
2008

Citing 10
Cited 0

Exploiting parallelism in pattern matching: an information retrieval application

ACM Transactions on Information Systems (TOIS)
A complete guide to DB2 universal database

A complete guide to DB2 universal database
Active Disks for Large-Scale Data Processing

Computer
VLSI Accelerators for Large Database Systems

IEEE Micro
An Associative Accelerator for Large Databases

IEEE Micro
A Fine-Grain Architecture for Relational Database Aggregation Operations

IEEE Micro
DiscoveryLink: a system for integrated access to life sciences data sources

IBM Systems Journal - Deep computing for the life sciences
Efficient Data Access for Parallel BLAST

IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Papers - Volume 01
MapReduce: simplified data processing on large clusters

OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6
Overview of the Blue Gene/L system architecture

IBM Journal of Research and Development

Quantified Score

Hi-index	0.00

Visualization

Abstract

We describe a software framework for deploying, scheduling and executing parallel DBMS user-defined programs on an attached high-performance computer (HPC) platform. This framework is advantageous for many DBMS workloads in the following two aspects. First, the long-running user-defined programs can be speeded up by taking advantage of the greater hardware parallel-ism available on the attached HPC platform. Second, the interac-tive response time of the remaining applications on the database server platform is improved by the off-loading of long-running user-defined programs to the attached HPC platform. Our frame-work provides a new approach for integrating high-performance computing into the workflow of query-oriented, computationally-intensive applications.