-
Transitioning from file-based HPC workflows to streaming data pipelines with openPMD and ADIOS2
Authors:
Franz Poeschel,
Juncheng E,
William F. Godoy,
Norbert Podhorszki,
Scott Klasky,
Greg Eisenhauer,
Philip E. Davis,
Lipeng Wan,
Ana Gainaru,
Junmin Gu,
Fabian Koller,
René Widera,
Michael Bussmann,
Axel Huebl
Abstract:
This paper aims to create a transition path from file-based IO to streaming-based workflows for scientific applications in an HPC environment. By using the openPMP-api, traditional workflows limited by filesystem bottlenecks can be overcome and flexibly extended for in situ analysis. The openPMD-api is a library for the description of scientific data according to the Open Standard for Particle-Mes…
▽ More
This paper aims to create a transition path from file-based IO to streaming-based workflows for scientific applications in an HPC environment. By using the openPMP-api, traditional workflows limited by filesystem bottlenecks can be overcome and flexibly extended for in situ analysis. The openPMD-api is a library for the description of scientific data according to the Open Standard for Particle-Mesh Data (openPMD). Its approach towards recent challenges posed by hardware heterogeneity lies in the decoupling of data description in domain sciences, such as plasma physics simulations, from concrete implementations in hardware and IO. The streaming backend is provided by the ADIOS2 framework, developed at Oak Ridge National Laboratory. This paper surveys two openPMD-based loosely-coupled setups to demonstrate flexible applicability and to evaluate performance. In loose coupling, as opposed to tight coupling, two (or more) applications are executed separately, e.g. in individual MPI contexts, yet cooperate by exchanging data. This way, a streaming-based workflow allows for standalone codes instead of tightly-coupled plugins, using a unified streaming-aware API and leveraging high-speed communication infrastructure available in modern compute clusters for massive data exchange. We determine new challenges in resource allocation and in the need of strategies for a flexible data distribution, demonstrating their influence on efficiency and scaling on the Summit compute system. The presented setups show the potential for a more flexible use of compute resources brought by streaming IO as well as the ability to increase throughput by avoiding filesystem bottlenecks.
△ Less
Submitted 19 January, 2022; v1 submitted 13 July, 2021;
originally announced July 2021.
-
Search for flares and associated CMEs on late-type main-sequence stars in optical SDSS spectra
Authors:
Florian Koller,
Martin Leitzinger,
Manuela Temmer,
Petra Odert,
Paul G. Beck,
Astrid Veronig
Abstract:
This work aims to detect and classify stellar flares and potential stellar coronal mass ejection (CME) signatures in optical spectra provided by the Sloan Digital Sky Survey (SDSS) data release 14. The sample is constrained to all F, G, K, and M main-sequence type stars, resulting in more than 630,000 stars. This work makes use of the individual spectral exposures provided by the SDSS.
An automa…
▽ More
This work aims to detect and classify stellar flares and potential stellar coronal mass ejection (CME) signatures in optical spectra provided by the Sloan Digital Sky Survey (SDSS) data release 14. The sample is constrained to all F, G, K, and M main-sequence type stars, resulting in more than 630,000 stars. This work makes use of the individual spectral exposures provided by the SDSS.
An automatic flare search was performed by detecting significant amplitude changes in the $Hα$ and $Hβ$ spectral lines after a Gaussian profile was fit to the line core. CMEs were searched for by identifying asymmetries in the Balmer lines caused by the Doppler effect of plasma motions in the line of sight.
We identified 281 flares on late-type stars (spectral types K3 to M9). We identified six possible CME candidates showing excess flux in Balmer line wings. Flare energies in $Hα$ were calculated and masses of the CME candidates were estimated. The derived $Hα$ flare energies range from $3 \times 10^{28} - 2 \times 10^{33}$ erg. The $Hα$ flare energy increases with earlier types, while the fraction of flaring times increases with later types. Mass estimates for the CME candidates are in the range of $6 \times 10^{16} - 6 \times 10^{18}$ g, and the highest projected velocities are $\sim300 - 700\:$ km/s.
The low detection rate of CMEs we obtained agrees with previous studies, suggesting that for late-type main-sequence stars the CME occurrence rate that can be detected with optical spectroscopy is low.
△ Less
Submitted 1 December, 2020;
originally announced December 2020.
-
A census of Coronal Mass Ejections on solar-like stars
Authors:
M. Leitzinger,
P. Odert,
R. Greimel,
K. Vida,
L. Kriskovics,
E. W. Guenther,
H. Korhonen,
F. Koller,
A. Hanslmeier,
Zs. Kővári,
H. Lammer
Abstract:
Coronal Mass Ejections (CMEs) may have major importance for planetary and stellar evolution. Stellar CME parameters, such as mass and velocity, have yet not been determined statistically. So far only a handful of stellar CMEs has been detected mainly on dMe stars using spectroscopic observations. We therefore aim for a statistical determination of CMEs of solar-like stars by using spectroscopic da…
▽ More
Coronal Mass Ejections (CMEs) may have major importance for planetary and stellar evolution. Stellar CME parameters, such as mass and velocity, have yet not been determined statistically. So far only a handful of stellar CMEs has been detected mainly on dMe stars using spectroscopic observations. We therefore aim for a statistical determination of CMEs of solar-like stars by using spectroscopic data from the ESO phase 3 and Polarbase archives. To identify stellar CMEs we use the Doppler signal in optical spectral lines being a signature of erupting filaments which are closely correlated to CMEs. We investigate more than 3700 hours of on-source time of in total 425 dF-dK stars. We find no signatures of CMEs and only few flares. To explain this low level of activity we derive upper limits for the non detections of CMEs and compare those with empirically modelled CME rates. To explain the low number of detected flares we adapt a flare power law derived from EUV data to the Hα regime, yielding more realistic results for Hα observations. In addition we examine the detectability of flares from the stars by extracting Sun-as-a-star Hα light curves. The extrapolated maximum numbers of observable CMEs are below the observationally determined upper limits, which indicates that the on-source times were mostly too short to detect stellar CMEs in Hα. We conclude that these non detections are related to observational biases in conjunction with a low level of activity of the investigated dF-dK stars.
△ Less
Submitted 11 February, 2020;
originally announced February 2020.
