170 likes | 187 Views
This paper discusses the implementation of a grid-enabled pattern matching system within the DAME e-Science Pilot Project. It explores the objectives of DAME, diagnostic issues, the role of AURA, and the AURA-G implementation. The paper concludes with the benefits and applicability of AURA-G in other domains.
E N D
Grid Enabled Pattern Matching within the DAME e-Science Pilot Project Jim Austin Computer Science University of York
Rolls-Royce University of Sheffield, P Fleming. University of Leeds, Peter Dew, Alison McKay. York, J Austin, J McDermid, A Wellings. University of Oxford, Lionel Tarassenko. Rolls-Royce, Derby. Data Systems and Solutions. Cybula Ltd. All hands 2002
Introduction • Objectives of DAME • Diagnostics issues • How AURA fits in • AURA-G – GRID enabled AURA • Where are we now? All hands 2002
DAME Objectives • DAME: Distributed Aircraft Maintenance Environment. • Demonstrate diagnostic capability on the GRID • Examine timeliness properties of the GRID • Demonstrate on the RR Aeroengine diagnostic problem All hands 2002
Engine flight data London Airport Airline office New York Airport Grid Diagnostics centre Maintenance Centre American data center European data center All hands 2002
Diagnostic issues • The system must analyse and report • Novel engine operation • Identify any cause of events • Do this quickly • Data • Large (many Tb) All hands 2002
Data – Zmod plots All hands 2002
Diagnostic station Engine data Novelty indication Quote Data used to identify novelty Data reduction processes Match requests Features Data to be searched for Proposed pattern matching process Data stores/ data warehouse Diagnosis AURA-G All hands 2002
How does AURA contribute • Search technology for multi-media data • Parallel pattern match engine based on neural networks. • Built on Correlation Matrix Memories. • High performance Beowulf and dedicated hardware implementations. • Commercially sold by Cybula Ltd. All hands 2002
Cortex-1 AURA parallel implementation 28 dedicated PCI based processors Beowulf configuration 3.5Gb memory size All hands 2002
inputs Basic CMM Samples of tracked orders All hands 2002
CMM Data sample DM coding Simple example of processing chain Matching previous events All hands 2002
Typical pre-processing Frequency 01101111011110111 DM coding (1 up 0 down) Time Fast Preserves information Produces a binary vector All hands 2002
Diagnostic station Engine data Novelty indication Quote Data used to identify novelty Data reduction processes Match requests Features Data to be searched for Pattern match results Data stores/ data warehouse Diagnosis AURA-G GRID All hands 2002
AURA-G • This is a Globus enabled AURA implementation. • Developed under DAME • Will be available end of 2002 for use in other problems. All hands 2002
AURA-G • Support of scalable pattern matching • Supports distributed search, across multiple CMM engines at different sites • OGSA compliant All hands 2002
Conclusions • AURA-G enabling fast access to large, complex data. • Available for other applications • Diagnostic framework in DAME applicable elsewhere. • DAME web site: www.cs.york.ac.uk/dame • AURA website: • www.cs.york.ac.uk/arch/nn/aura.html • www.cybula.com All hands 2002