Compact data structure and scalable algorithms for the sparse grid technique

Murarasu, Alin; Weidendorfer, Josef; Buse, Gerrit; Butnaru, Daniel; Pflüger, Dirk

doi:10.1145/1941553.1941559

Cited by 21 publications

(18 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the increase in dimensionality, the scalability of storing and retrieving (interpolating) simulation data becomes crucial to ensure interactive steering. Efficient algorithms performing the interpolation on accelerator cards such as GPGPUs are already under development [11], while dynamic extensions of the parameter ranges and automatic refinement based on user behavior and error measure need to be addressed in the near future. The quality of the visualization and user interaction are also critical for the success of any visual computational steering process, and thus in the focus of our project.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Towards High-Dimensional Computational Steering of Precomputed Simulation Data using Sparse Grids

Butnaru

Pflüger

Bungartz

2011

Procedia Computer Science

Self Cite

View full text Add to dashboard Cite

With the ever-increasing complexity, accuracy, dimensionality, and size of simulations, a step in the direction of data-intensive scientific discovery becomes necessary. Parameter-dependent simulations are an example of such a data-intensive tasks: The researcher, who is interested in the dependency of the simulation's result on a set of input parameters, changes essential parameters and wants to immediately see the effect of the changes in a visual environment. In this scenario, an interactive exploration is not possible due to the long execution time needed by even a single simulation corresponding to one parameter combination and the overall large number of parameter combinations which could be of interest.In this paper, we present a method for computational steering with pre-computed data as a particular form of visual scientific exploration. We consider a parametrized simulation as a multi-variate function in several parameters. Using the technique of sparse grids, this makes it possible to sample and compress potentially high-dimensional parameter spaces and to efficiently deliver a combination of simulated and precomputed data to the steering process, thus enabling the user to interactively explore high-dimensional simulation results.

show abstract

Section: Discussionmentioning

confidence: 99%

“…This can be executed very efficiently on accelerator cards or other parallel environments and guarantees a delivery of data to the visualization which is fast enough for interactive interaction [11].…”

Section: The Data Extraction Process: Interpolationmentioning

confidence: 99%

Towards High-Dimensional Computational Steering of Precomputed Simulation Data using Sparse Grids

Butnaru

Pflüger

Bungartz

2011

Procedia Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…• We compare the performance on 4 multi-core systems from Intel and AMD. Our input oriented optimizations result in a speedup factor of up to 9x compared to the state-of-the-art sparse grid interpolation from [4]. The OpenMP version scales linearly on all benchmarked multi-cores.…”

Section: Introductionmentioning

confidence: 99%

“…1. Highdimensional simulation data is compressed using a sparse grid operation referred to as hierarchization [4]. Afterwards, the simulation data is decompressed for real-time visualization using sparse grid interpolation or evaluation [4].…”

Section: Introductionmentioning

confidence: 99%

“…Sparse grids allow us to approximate d-dimensional functions and are typically stored in memory using trees and hash-tables. [4] proposes a bijection based data structure for sparse grids that consumes a minimum amount of memory. The main idea there is to store only the values in the sparse grid in a special order such that the position of a value can be passed to a bijective mapping in order to retrieve the coordinates of the d-dimensional point corresponding to that value.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Building Input Adaptive Parallel Applications: A Case Study of Sparse Grid Interpolation

Murarasu¹,

Weidendorfer²

2012

2012 IEEE 15th International Conference on Computational Science and Engineering

Self Cite

View full text Add to dashboard Cite

The well-known power wall resulting in multi-cores requires special techniques for speeding up applications. In this sense, parallelization plays a crucial role. Besides standard serial optimizations, techniques such as input specialization can also bring a substantial contribution to the speedup. By identifying common patterns in the input data, we propose new algorithms for sparse grid interpolation that accelerate the state-of-the-art non-specialized version. Sparse grid interpolation is an inherently hierarchical method of interpolation employed for example in computational steering applications for decompressing highdimensional simulation data. In this context, improving the speedup is essential for real-time visualization. Using input specialization, we report a speedup of up to 9x over the nonspecialized version. The paper covers the steps we took to reach this speedup by means of input adaptivity. Our algorithms will be integrated in fastsg, a library for fast sparse grid interpolation.

show abstract

A Cache-Optimal Alternative to the Unidirectional Hierarchization Algorithm

Hupp

Jacob

2016

Lecture Notes in Computational Science and Engineering

View full text Add to dashboard Cite

Compact data structure and scalable algorithms for the sparse grid technique

Cited by 21 publications

References 8 publications

Towards High-Dimensional Computational Steering of Precomputed Simulation Data using Sparse Grids

Towards High-Dimensional Computational Steering of Precomputed Simulation Data using Sparse Grids

Building Input Adaptive Parallel Applications: A Case Study of Sparse Grid Interpolation

A Cache-Optimal Alternative to the Unidirectional Hierarchization Algorithm

Contact Info

Product

Resources

About