In this bimonthly feature, HPCwire highlights newly published research in the high-performance computing community and related domains. From parallel programming to exascale to quantum computing, the details are here.
Running green HPC simulations of a mosquito
A warming climate has enabled some major disease vectors, like mosquitoes, to spread more broadly and through more of the year. These authors – a team from the National University of La Plata and the Autonomous University of Barcelona – use GPU-powered high-performance simulations to evaluate the reproduction and evolution of the Aedes aegypti mosquito, which is the predominant vector for dengue fever, Zika and the chikungunya virus.
Authors: Erica Montes de Oca, Remo Suppi, Laura De Giusti and Marcelo Naiouf.
Studying energy efficiency at SCAYLE
As the exascale era looms, the massive power requirements of supercomputing centers are a growing concern. These authors from the Supercomputing Centre of Castile and Leon (SCAYLE) in Spain examine the energy efficiency of the supercomputing center, highlighting the role of weather conditions, density and more in ensuring power-efficient operation of large datacenters.
Authors: A. Fernández González, V. Matellán, J. M. Martínez García, J. Lorenzana and M. López.
Completing “the Last Journey” on Mira
The Last Journey, these authors (a team from Argonne National Laboratory and the University of Chicago) explain, is a large-volume, gravity-only, cosmological N-body simulation involving over 1.24 trillion particles. The authors discuss running this massive simulation using the HACC simulation and analysis framework on the Mira supercomputer at Argonne National Laboratory, highlighting how they managed the resulting data and refined the simulation.
Authors: Katrin Heitmann, Nicholas Frontiere, Esteban Rangel, Patricia Larsen, Adrian Pope, Imran Sultan, Thomas Uram, Salman Habib, Hal Finkel, Danila Korytov, Eve Kovacs, Silvio Rizzi and Joe Insley.
Addressing idle time on Folding@home during the pandemic
Crowdsourced computing leader Folding@home amassed an enormous army of volunteer systems as it crunched the viral proteins of COVID-19 – but it may have been off more than it could chew. In this paper, Nane Kratzke of the Lübeck University of Applied Sciences discusses how Folding@home’s inability to share its resources has led to substantial idle times. Kratzke highlights various existing technologies – such as containers and serverless architectures – that could be used to share this idle time with other researchers.
Author: Nane Kratzke
Analyzing the trends in Top500 leaders and Gordon Bell Prize recipients
A top rank in the Top500 and the Gordon Bell Prize may be the two biggest honors in the supercomputing sector. This author, Weimin Zheng of Tsinghua University, reviews the supercomputers in the Top500 list and the last seven Gordon Bell Prize recipients to examine trends in the cutting edge of supercomputing. Zheng highlights a trend toward heterogeneous architectures, AI applications and increasing difficulty in scientific simulations under a more heterogeneous status quo.
Author: Weimin Zheng
Preparing for star-by-star simulation in the exascale era
Exascale computing will allow astronomers to perform star-by-star simulations of galaxies, which will be a game-changer for analyses of star and galaxy formation. In this paper, a team from Riken, the University of Tokyo, Kobe University and the Tokyo Institute of Technology discuss how computational models in astronomy will need to change to fit into this new era, proposing a new stochastic star formation model for star-by-star simulations.
Authors: Yutaka Hirai, Michiko S. Fujii and Takayuki R. Saitoh.
Balancing power and performance in HPC clouds
Cloud HPC is often presented as a low-footprint alternative to on-premises systems – however, cloud systems have grown in size to the point where their energy consumption is a serious consideration. These authors present a mechanism combining a virtual machine placement algorithm with a frequency scaling method to reduce energy consumption of cloud HPC environments with only a “modest” impact on performance.
Authors: Lixia Chen, Jian Li, Ruhui Ma, Haibing Guan and Hans-Arno Jacobsen.
Do you know about research that should be included in next month’s list? If so, send us an email at [email protected]. We look forward to hearing from you.