Montblanc: GPU accelerated Radio Interferometer Measurement Equations in support of Bayesian Inference for Radio Observations

We present Montblanc, a GPU implementation of the Radio interferometer measurement equation (RIME) in support of the Bayesian inference for radio observations (BIRO) technique. BIRO uses Bayesian inference to select sky models that best match the visibilities observed by a radio interferometer. To accomplish this, BIRO evaluates the RIME multiple times, varying sky model parameters to produce multiple model visibilities. Chi-squared values computed from the model and observed visibilities are used as likelihood values to drive the Bayesian sampling process and select the best sky model. As most of the elements of the RIME and chi-squared calculation are independent of one another, they are highly amenable to parallel computation. Additionally, Montblanc caters for iterative RIME evaluation to produce multiple chi-squared values. Modified model parameters are transferred to the GPU between each iteration. We implemented Montblanc as a Python package based upon NVIDIA's CUDA architecture. As such, it is easy to extend and implement different pipelines. At present, Montblanc supports point and Gaussian morphologies, but is designed for easy addition of new source profiles. Montblanc's RIME implementation is performant: On an NVIDIA K40, it is approximately 250 times faster than MeqTrees on a dual hexacore Intel E5-2620v2 CPU. Compared to the OSKAR simulator's GPU-implemented RIME components it is 7.7 and 12 times faster on the same K40 for single and double-precision floating point respectively. However, OSKAR's RIME implementation is more general than Montblanc's BIRO-tailored RIME. Theoretical analysis of Montblanc's dominant CUDA kernel suggests that it is memory bound. In practice, profiling shows that is balanced between compute and memory, as much of the data required by the problem is retained in L1 and L2 cache.

[1]  Kenneth Levenberg A METHOD FOR THE SOLUTION OF CERTAIN NON – LINEAR PROBLEMS IN LEAST SQUARES , 1944 .

[2]  Jan Noordam,et al.  Radio Interferometric Calibration Using The SAGE Algorithm , 2008, DSP 2009.

[3]  Tim J. Cornwell,et al.  The Noncoplanar Baselines Effect in Radio Interferometry: The W-Projection Algorithm , 2008, IEEE Journal of Selected Topics in Signal Processing.

[4]  David J. C. MacKay,et al.  Information Theory, Inference, and Learning Algorithms , 2004, IEEE Transactions on Information Theory.

[5]  Jean-Luc Starck,et al.  Astronomical Data Analysis , 2007 .

[6]  Francesc Alted,et al.  Why Modern CPUs Are Starving and What Can Be Done about It , 2010, Computing in Science & Engineering.

[7]  Kim M. Hazelwood,et al.  Where is the data? Why you cannot debate CPU vs. GPU performance without the answer , 2011, (IEEE ISPASS) IEEE INTERNATIONAL SYMPOSIUM ON PERFORMANCE ANALYSIS OF SYSTEMS AND SOFTWARE.

[8]  Christopher J. Fluke,et al.  Analysing astronomy algorithms for graphics processing units and beyond: Analysing algorithms for GPUs and beyond , 2010 .

[9]  O. Smirnov Revisiting the radio interferometer measurement equation. I. A full-sky Jones formalism , 2011, 1101.1764.

[10]  Stephanie Thalberg,et al.  Interferometry And Synthesis In Radio Astronomy , 2016 .

[11]  E. Bertin,et al.  SExtractor: Software for source extraction , 1996 .

[12]  Mark J. Harris Mapping computational concepts to GPUs , 2005, SIGGRAPH Courses.

[13]  R. Sault,et al.  Understanding radio polarimetry. I. Mathematical foundations , 1996 .

[14]  Oleg M. Smirnov,et al.  Revisiting the radio interferometer measurement equation. II. Calibration and direction-dependent effects , 2011, 1101.1765.

[15]  G.E. Moore,et al.  Cramming More Components Onto Integrated Circuits , 1998, Proceedings of the IEEE.

[16]  J. Högbom,et al.  APERTURE SYNTHESIS WITH A NON-REGULAR DISTRIBUTION OF INTERFEROMETER BASELINES. Commentary , 1974 .

[17]  O. Smirnov,et al.  The MeqTrees software system and its use for third-generation calibration of radio interferometers , 2010, 1101.1745.

[18]  F. Feroz,et al.  Bayesian modelling of clusters of galaxies from multifrequency‐pointed Sunyaev–Zel'dovich observations , 2008, 0811.1199.

[19]  Michael J. Flynn,et al.  Some Computer Organizations and Their Effectiveness , 1972, IEEE Transactions on Computers.

[20]  F. Feroz,et al.  MultiNest: an efficient and robust Bayesian inference tool for cosmology and particle physics , 2008, 0809.3437.

[21]  Attila Popping,et al.  The standing wave phenomenon in radio telescopes - Frequency modulation of the WSRT primary beam , 2007, 0712.2303.

[22]  Ivan R. King,et al.  The structure of star clusters. III. Some simple dvriamical models , 1966 .

[23]  Nicolas Pinto,et al.  PyCUDA and PyOpenCL: A scripting-based approach to GPU run-time code generation , 2009, Parallel Comput..

[24]  Klaus Maisinger,et al.  Maximum-likelihood estimation of the cosmic microwave background power spectrum from interferometer observations , 2002 .

[25]  Gaël Varoquaux,et al.  The NumPy Array: A Structure for Efficient Numerical Computation , 2011, Computing in Science & Engineering.

[26]  William Kahan,et al.  Pracniques: further remarks on reducing truncation errors , 1965, CACM.

[27]  J. Sérsic,et al.  Influence of the atmospheric and instrumental dispersion on the brightness distribution in a galaxy , 1963 .

[28]  F. R. Harnden,et al.  Astronomical Data Analysis Software and Systems X , 2001 .

[29]  M. Kunz,et al.  Bayesian inference for radio observations , 2015, 1501.05304.

[30]  O. M. Smirnov,et al.  Revisiting the radio interferometer measurement equation , 2011 .

[31]  S. White,et al.  A Universal Density Profile from Hierarchical Clustering , 1996, astro-ph/9611107.

[32]  Ivan R. King,et al.  Density Data and Emission Measure for a Model of the Coma Cluster , 1972 .

[33]  D. Marquardt An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[34]  Stefano Salvini,et al.  OSKAR: Simulating digital beamforming for the SKA aperture array , 2010, 2010 IEEE International Symposium on Phased Array Systems and Technology.

[35]  P. S. Bunclark,et al.  Astronomical Data Analysis Software and Systems , 2008 .