Virtual data space – load balancing for irregular applications

Decker, Thomas

doi:10.1016/s0167-8191(00)00061-2

“…The first issue introduces two well known and difficult problems: (a) load balancing and (b) fault-tolerance. The scientific computing community (including our group) has developed applicationspecific load balancing libraries and general purpose software runtime systems [Basermann et al 2000;Yuan et al 1997;Devine et al 2000;Decker 2000;Friedman et al 1997;Kalé and Krishnan 1993;Kohn and Baden 2001;Barker et al 2004]. The issue of fault tolerance for parallel and distributed computing has also been addressed previously [Fagg and Dongarra 2000;Sankaran et al 2005;Chakravorty and Kalé 2004;Elnozahy et al 2002].…”

Section: Introductionmentioning

confidence: 99%

Toward Real-Time Image Guided Neurosurgery Using Distributed and Grid Computing

Chrisochoides

¹

,

Fedorov

²

,

Kot

³

et al. 2006

ACM/IEEE SC 2006 Conference (SC'06)

View full text Add to dashboard Cite

Neurosurgical resection is a therapeutic intervention in the treatment of brain tumors. Precision of the resection can be improved by utilizing Magnetic Resonance Imaging (MRI) as an aid in decision making during Image Guided Neurosurgery (IGNS). Image registration adjusts pre-operative data according to intra-operative tissue deformation. Some of the approaches increase the registration accuracy by tracking image landmarks through the whole brain volume. High computational cost used to render these techniques inappropriate for clinical applications.In this paper we present a parallel implementation of a state of the art registration method, and a number of needed incremental improvements. Overall, we reduced the response time for registration of an average dataset from about an hour and for some cases more than an hour to less than seven minutes, which is within the time constraints imposed by neurosurgeons. For the first time in clinical practice we demonstrated, that with the help of distributed computing non-rigid MRI registration based on volume tracking can be computed intra-operatively.

show abstract

“…The first issue introduces two well known and difficult problems: (a) load balancing and (b) fault-tolerance. The scientific computing community (including our group) has developed applicationspecific load balancing libraries and general purpose software runtime systems [Basermann et al 2000;Yuan et al 1997;Devine et al 2000;Decker 2000;Friedman et al 1997;Kalé and Krishnan 1993;Kohn and Baden 2001;Barker et al 2004]. The issue of fault tolerance for parallel and distributed computing has also been addressed previously [Fagg and Dongarra 2000;Sankaran et al 2005;Chakravorty and Kalé 2004;Elnozahy et al 2002].…”

Section: Introductionmentioning

confidence: 99%

Imaging and visual analysis---Toward real-time image guided neurosurgery using distributed and grid computing

Chrisochoides

¹

,

Fedorov

²

,

Kot

³

et al. 2006

Proceedings of the 2006 ACM/IEEE Conference on Supercomputing - SC '06

17

0

View full text Add to dashboard Cite

Neurosurgical resection is a therapeutic intervention in the treatment of brain tumors. Precision of the resection can be improved by utilizing Magnetic Resonance Imaging (MRI) as an aid in decision making during Image Guided Neurosurgery (IGNS). Image registration adjusts pre-operative data according to intra-operative tissue deformation. Some of the approaches increase the registration accuracy by tracking image landmarks through the whole brain volume. High computational cost used to render these techniques inappropriate for clinical applications.In this paper we present a parallel implementation of a state of the art registration method, and a number of needed incremental improvements. Overall, we reduced the response time for registration of an average dataset from about an hour and for some cases more than an hour to less than seven minutes, which is within the time constraints imposed by neurosurgeons. For the first time in clinical practice we demonstrated, that with the help of distributed computing non-rigid MRI registration based on volume tracking can be computed intra-operatively.

show abstract

“…In [17], Feldmann, Mysliwietz, and Monien investigate the behavior of parallel MIN/MAX-tree evaluation in the context of parallel game (chess) programs employing work-stealing strategies. In [14], Decker introduces VDS (virtual data space), a load-balancing system for irregular applications that makes use of work-stealing and other strategies. In [23], Mahapatra and Dutt use work-stealing for parallel branch and bound algorithms.…”

Section: Introductionmentioning

confidence: 99%

The Natural Work-Stealing Algorithm is Stable

Berenbrink

¹

,

Friedetzky

²

,

Goldberg

³

2003

View full text Add to dashboard Cite

The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that: • a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.

show abstract

Virtual data space – load balancing for irregular applications

Cited by 10 publications

References 42 publications

Toward Real-Time Image Guided Neurosurgery Using Distributed and Grid Computing

Toward Real-Time Image Guided Neurosurgery Using Distributed and Grid Computing

Imaging and visual analysis---Toward real-time image guided neurosurgery using distributed and grid computing

The Natural Work-Stealing Algorithm is Stable

Contact Info

Product

Resources

About