The Sarnoff Engine: a massively parallel computer for high definition system simulation

Knight, S.; Chin, D.; Taylor, Hebron; Peters, J.

doi:10.1109/asap.1992.218560

Cited by 5 publications

(2 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is quite distinct from other so-called massively parallel systems, such as the Intel Touchstone and Paragon, MasPar MPI/2, CM2/5, Kendall Square KSR1 or IBM PVS systems in that it was conceived and designed for continuous real-time signal processing, and in that it offers an incomparably superior massively-parallel, multiprocessor high performance computational environment multi-processor for voxel operations processing. [44][45][46][47][48][49][50][51][52] It is first described in 4s and later in. 49 Reference 5~ provides the most recent description.…”

Section: Anatomically Photorealistic Rendering and Scenario Simulationmentioning

confidence: 99%

Interventional telemedicine for noninvasive neuroradiosurgery: Remote-site high-performance computing, mathematical optimization, and virtual scenario simulation

et al. 1995

View full text Add to dashboard Cite

Interventional Telemedicine may have potential utility in providing connectivity and access to specialized high performance computing and advanced software resources in support of clinical procedures in the field of Minimally-Invasive Surgery and Stereotactic Linear Accelerator (LINAC) Radiosurgery. Such interventions may benefit from application of nonlinear quadratic inverse solution methods designed to provide the capability to reverse optimize a 'best case' treatment plan. The formidable decision-making challenges posed by increasingly complex optimized data and progressively versatile LINAC delivery systems require volume visualization of projected treatment data and imaging anatomy via photorealistic rendering and virtual scenario simulation techniques. Both these new directions are heavily dependent on access to specialized high performance computing platforms solely accessible via broad-bandwidth network connectivity. This pilot project presents resimulation of retrospective radiosurgical case data using inverse solution optimization models running on workstation clusters and then volume rendered and simulated on the Princeton Graphics Engine Supercomputer. Evidence for effective utilization of such optimization and virtual simulation methods running on remotely accessed, distant high performance computing resources is discussed in view of the potential for long-term clinical investigation and eventual development of Interventional Telemedicine as a clinically practical approach for providing support to remote or non-urban radiosurgery centers in the industrialized and developing world.

show abstract

Section: Anatomically Photorealistic Rendering and Scenario Simulationmentioning

confidence: 99%

Interventional telemedicine for noninvasive neuroradiosurgery: Remote-site high-performance computing, mathematical optimization, and virtual scenario simulation

et al. 1995

View full text Add to dashboard Cite

show abstract

“…It is quite straightforward to implement PART on an SIMD linear processor array such as the SliM-II [16] or the Sarnoff Engine [17]. Another potential implementation can utilize a linear array of PC's, each connected to its two neighbors by dedicated high-bandwidth communication cards.…”

Section: Introductionmentioning

confidence: 99%

Parallel ART for image reconstruction in CT using processor arrays

Gordon

2006

International Journal of Parallel, Emergent and Distributed Sys

View full text Add to dashboard Cite

Algebraic Reconstruction Technique (ART) is a widely-used iterative method for solving sparse systems of linear equations. This method (originally due to Kaczmarz) is inherently sequential according to its mathematical definition since, at each step, the current iterate is projected toward one of the hyperplanes defined by the equations. The main advantages of ART are its robustness, its cyclic convergence on inconsistent systems, and its relatively good initial convergence. ART is widely used as an iterative solution to the problem of image reconstruction from projections in computerized tomography (CT), where its implementation with a small relaxation parameter produces excellent results. It is shown that for this particular problem, ART can be implemented in parallel on a linear processor array. Reconstructing an image of n pixels from Q(n) equations can be done on a linear array of p ¼ Oð ffiffi ffi n p Þ processors with optimal efficiency (linear speedup) and O(n/p) memory for each processor. The parallel technique can be applied to various geometric models of image reconstruction, as well as to 3D reconstruction with spherically symmetric volume elements, using a 2D rectangular mesh-connected array of processors.

show abstract

Efficient image processing algorithms on the scan line array processor

Helman

JáJá

1995

IEEE Trans. Pattern Anal. Machine Intell.

View full text Add to dashboard Cite

The Sarnoff Engine: a massively parallel computer for high definition system simulation

Cited by 5 publications

References 2 publications

Interventional telemedicine for noninvasive neuroradiosurgery: Remote-site high-performance computing, mathematical optimization, and virtual scenario simulation

Interventional telemedicine for noninvasive neuroradiosurgery: Remote-site high-performance computing, mathematical optimization, and virtual scenario simulation

Parallel ART for image reconstruction in CT using processor arrays

Efficient image processing algorithms on the scan line array processor

Contact Info

Product

Resources

About