Achieving low-cost high-reliability computation through redundant parallel processing

This paper presents a reconfigurable parallel architecture comprising an FPGA backbone and multiple processing nodes connected in a redundant array architecture and constructed mainly from low-cost commercial components. The reconfigurability of the backbone aids in allowing the system to operate as a fault-tolerant cluster utilising the principle of reliability through redundancy. Although initially designed for space-borne on-board processing of satellite imagery, the system combines the advantages of powerful computational resources with simplified software development. This makes the computer a useful general-purpose embedded processing block for critical computational tasks where fault-tolerance and high processing capabilities are required.

[1]  Ian McLoughlin,et al.  Embedded Linux platform for a fault tolerant space based parallel computer , 2004 .

[2]  Ian McLoughlin,et al.  Error Detection and Correction for microsatellite software running on {COTS} processors , 2001 .

[3]  James R. Wertz,et al.  Space Mission Analysis and Design , 1992 .

[4]  Ian McLoughlin,et al.  Fault tolerance through redundant COTS components for satellite processing applications , 2003, Fourth International Conference on Information, Communications and Signal Processing, 2003 and the Fourth Pacific Rim Conference on Multimedia. Proceedings of the 2003 Joint.

[5]  Ian McLoughlin,et al.  Reconfigurable, fault tolerant and high performance payload for space missions , 2003 .

[6]  Liu Bo,et al.  D-ISMC: a distributed unsupervised classification algorithm for optical satellite imagery , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[7]  Ian McLoughlin.  Design, testing and verification of a microsatellite on-board data processing unit using commercial grade processors , 2001 .

[8]  Ian McLoughlin,et al.  Centralised computation service architecture for the X-Sat micro-satellite , 2005 .

[9]  Ian McLoughlin,et al.  First beowulf cluster in space , 2005 .

[10]  Guillermo Ortega.  Linux for the International Space Station Program , 1999 .

[11]  Thomas Sterling,et al.  An assessment of Beowulf-class computing for NASA requirements: initial findings from the first NASA workshop on Beowulf-class clustered computing , 1998, 1998 IEEE Aerospace Conference Proceedings (Cat. No.98TH8339).