Scikit-Cuda 0.5.1

Givon, Lev E.; Schlüter, Jan; Erichson, N. Benjamin; Walt, Stéfan van der; Dieleman, Sander; Bastien, Frédéric; Larson, Eric B.; Unterthiner, Thomas; Shi, Xing; Pfister, Luke A.; Capdevila, Chris; Moldovan, Teodor Mihai; Lee, Gregory R.; Thomas, Brian G.; Rubinsteyn, Alex; Chiang, David Wei

doi:10.5281/zenodo.40565

Cited by 3 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The original CPU code was written in Python for easy integration with other data acquisition, analysis, and visualization tools provided at NSLS-II. Therefore, in our port to NVIDIA GPUs we use PyCUDA (wrapper for CUDA API) [19], scikit-cuda (for cuFFT and other CUDA libraries) [20] and MPI for Python (mpi4py) [21]- [23] to accelerate the existing Python code. In addition, most of the computation are rewritten as CUDA C kernels, which are called through the PyCUDA binding for the best performance.…”

Section: Detailsmentioning

confidence: 99%

High-Performance Multi-Mode Ptychography Reconstruction on Distributed GPUs

Dong¹,

Fang²,

Huang³

et al. 2018

2018 New York Scientific Data Summit (NYSDS)

View full text Add to dashboard Cite

Ptychography is an emerging imaging technique that is able to provide wavelength-limited spatial resolution from specimen with extended lateral dimensions. As a scanning microscopy method, a typical two-dimensional image requires a number of data frames. As a diffraction-based imaging technique, the realspace image has to be recovered through iterative reconstruction algorithms. Due to these two inherent aspects, a ptychographic reconstruction is generally a computation-intensive and timeconsuming process, which limits the throughput of this method. We report an accelerated version of the multi-mode difference map algorithm for ptychography reconstruction using multiple distributed GPUs. This approach leverages available scientific computing packages in Python, including mpi4py and PyCUDA, with the core computation functions implemented in CUDA C. We find that interestingly even with MPI collective communications, the weak scaling in the number of GPU nodes can still remain nearly constant. Most importantly, for realistic diffraction measurements, we observe a speedup ranging from a factor of 10 to 10 3 depending on the data size, which reduces the reconstruction time remarkably from hours to typically about 1 minute and is thus critical for real-time data processing and visualization.

show abstract

Section: Detailsmentioning

confidence: 99%

High-Performance Multi-Mode Ptychography Reconstruction on Distributed GPUs

Dong¹,

Fang²,

Huang³

et al. 2018

2018 New York Scientific Data Summit (NYSDS)

View full text Add to dashboard Cite

show abstract

“…All matrix operations are conducted via the Numpy package (compiled against efficient matrix libraries). For the GPU implementation, we make use of the scikit-cuda package [10] and of custom CUDA kernels, called from Python via the PyCuda package [13].…”

Section: Setupmentioning

confidence: 99%

Massively-parallel break detection for satellite data

Mehren

Gieseke

Verbesselt

et al. 2018

Proceedings of the 30th International Conference on Scientific and Statistical Database Management

View full text Add to dashboard Cite

The field of remote sensing is nowadays faced with huge amounts of data. While this offers a variety of exciting research opportunities, it also yields significant challenges regarding both computation time and space requirements. In practice, the sheer data volumes render existing approaches too slow for processing and analyzing all the available data. This work aims at accelerating BFAST, one of the state-of-the-art methods for break detection given satellite image time series. In particular, we propose a massively-parallel implementation for BFAST that can effectively make use of modern parallel compute devices such as GPUs. Our experimental evaluation shows that the proposed GPU implementation is up to four orders of magnitude faster than the existing publicly available implementation and up to ten times faster than a corresponding multi-threaded CPU execution. The dramatic decrease in running time renders the analysis of significantly larger datasets possible in seconds or minutes instead of hours or days. We demonstrate the practical benefits of our implementations given both artificial and real datasets.

show abstract

“…In figs. 8 and 9, the performance of Bifrost is compared over various parameter settings with a serial pipeline implemented with the package scikit-cuda 20 (Givon et al, 2015). Both scikit-cuda and Bifrost have Python frontends, wrapping a cuFFT back end, meaning that differences in performance should measure the speedup due to pipeline parallelism, or slowness due to Python overhead.…”

Section: Performancementioning

confidence: 99%

Bifrost: A Python/C++ Framework for High-Throughput Stream Processing in Astronomy

Cranmer

Barsdell

Price

et al. 2017

J. Astron. Instrum.

View full text Add to dashboard Cite

Radio astronomy observatories with high throughput back end instruments require real-time data processing. While computing hardware continues to advance rapidly, development of real-time processing pipelines remains difficult and time-consuming, which can limit scientific productivity. Motivated by this, we have developed Bifrost: an open-source software framework for rapid pipeline development 1 . Bifrost combines a high-level Python interface with highly efficient reconfigurable data transport and a library of computing blocks for CPU and GPU processing. The framework is generalizable, but initially it emphasizes the needs of high-throughput radio astronomy pipelines, such as the ability to process data buffers as if they were continuous streams, the capacity to partition processing into distinct data sequences (e.g., separate observations), and the ability to extract specific intervals from buffered data. Computing blocks in the library are designed for applications such as interferometry, pulsar dedispersion and timing, and transient search pipelines. We describe the design and implementation of the Bifrost framework and demonstrate its use as the backbone in the correlation and beamforming back end of the Long Wavelength Array station in the Sevilleta National Wildlife Refuge, NM.

show abstract

Scikit-Cuda 0.5.1

Cited by 3 publications

References 0 publications

High-Performance Multi-Mode Ptychography Reconstruction on Distributed GPUs

High-Performance Multi-Mode Ptychography Reconstruction on Distributed GPUs

Massively-parallel break detection for satellite data

Bifrost: A Python/C++ Framework for High-Throughput Stream Processing in Astronomy

Contact Info

Product

Resources

About