☆ 4.2 Article

Reflecting on the Goal and Baseline for Exascale Computing: A Roadmap Based on Weather and Climate Simulations

COMPUTING IN SCIENCE & ENGINEERING (2019)

Journal

COMPUTING IN SCIENCE & ENGINEERING

Volume 21, Issue 1, Pages 30-41

Publisher

IEEE COMPUTER SOC

DOI: 10.1109/MCSE.2018.2888788

Keywords

-

Categories

Computer Science, Interdisciplinary Applications

Ask authors/readers for more resources

Protocol

Community support

Reagent

Community support

Abstract

We present a roadmap towards exascale computing based on true application performance goals. It is based on two state-of-the art European numerical weather prediction models (IFS from ECMWF and COSMO from MeteoSwiss) and their current performance when run at very high spatial resolution on present-day supercomputers. We conclude that these models execute about 100-250 times too slow for operational throughput rates at a horizontal resolution of 1 km, even when executed on a full petascale system with nearly 5000 state-of-the-art hybrid GPU-CPU nodes. Our analysis of the performance in terms of a metric that assesses the efficiency of memory use shows a path to improve the performance of hardware and software in order to meet operational requirements early next decade.

Authors

I am an author on this paper

Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.2

Not enough ratings

Secondary Ratings

Novelty

-

Significance

-

Scientific rigor

-

Rate this paper

Recommended

Article Computer Science, Hardware & Architecture

Making Intelligent Reflecting Surfaces More Intelligent: A Roadmap Through Reservoir Computing

Zhou Zhou, Kangjun Bai, Nima Mohammadi, Yang Yi, Lingjia Liu

Summary: This article introduces a neural network-based signal processing framework for intelligent reflecting surface-aided wireless communications systems, and discusses several practical issues.

IEEE NETWORK (2022)

Add to Collection

Article Engineering, Civil

EQSIM-A multidisciplinary framework for fault-to-structure earthquake simulations on exascale computers part I: Computational models and workflow

David McCallen, Anders Petersson, Arthur Rodgers, Arben Pitarka, Mamun Miah, Floriana Petrone, Bjorn Sjogreen, Norman Abrahamson, Houjun Tang

Summary: Computational simulations play a central role in seismic analysis and design of major infrastructure, now used to virtually test earthquake-induced response of structures. Advancements in high-performance computing have allowed simulations to increase understanding and prediction of ground motions for earthquake hazard assessments. Yet, the extreme computational requirements for regional-scale ground motion simulations present challenges in achieving higher frequency resolutions and exploring a large parametric space through direct simulations.

EARTHQUAKE SPECTRA (2021)

Add to Collection

Article Materials Science, Multidisciplinary

Roadmap on electronic structure codes in the exascale era

Vikram Gavini, Stefano Baroni, Volker Blum, David R. Bowler, Alexander Buccheri, James R. Chelikowsky, Sambit Das, William Dawson, Pietro Delugas, Mehmet Dogan, Claudia Draxl, Giulia Galli, Luigi Genovese, Paolo Giannozzi, Matteo Giantomassi, Xavier Gonze, Marco Govoni, Francois Gygi, Andris Gulans, John M. Herbert, Sebastian Kokott, Thomas D. Kuehne, Kai-Hsin Liou, Tsuyoshi Miyazaki, Phani Motamarri, Ayako Nakata, John E. Pask, Christian Plessl, Laura E. Ratcliff, Ryan M. Richard, Mariana Rossi, Robert Schade, Matthias Scheffler, Ole Schutt, Phanish Suryanarayana, Marc Torrent, Lionel Truflandier, Theresa L. Windus, Qimen Xu, Victor W-Z Yu, D. Perez

Summary: Electronic structure calculations provide important insights into physical and chemical properties of various systems, and they are crucial in fields like materials science and computational chemistry. As we enter the exascale era, there are new opportunities and challenges in terms of simulation capabilities and efficient use of new architectures. This roadmap discusses the current status, development priorities, and plans of 14 electronic structure codes in tackling these challenges and leveraging the benefits of exascale computing.

MODELLING AND SIMULATION IN MATERIALS SCIENCE AND ENGINEERING (2023)

Add to Collection

Article Computer Science, Theory & Methods

The EU Center of Excellence for Exascale in Solid Earth (ChEESE): Implementation, results, and roadmap for the second phase

Arnau Folch, Claudia Abril, Michael Afanasiev, Giorgio Amati, Michael Bader, Rosa M. Badia, Hafize B. Bayraktar, Sara Barsotti, Roberto Basili, Fabrizio Bernardi, Christian Boehm, Beatriz Brizuela, Federico Brogi, Eduardo Cabrera, Emanuele Casarotti, Manuel J. Castro, Matteo Cerminara, Antonella Cirella, Alexey Cheptsov, Javier Conejero, Antonio Costa, Marc de la Asuncion, Josep de la Puente, Marco Djuric, Ravil Dorozhinskii, Gabriela Espinosa, Tomaso Esposti-Ongaro, Joan Farnos, Nathalie Favretto-Cristini, Andreas Fichtner, Alexandre Fournier, Alice-Agnes Gabriel, Jean-Matthieu Gallard, Steven J. Gibbons, Sylfest Glimsdal, Jose Manuel Gonzalez-Vida, Jose Gracia, Rose Gregorio, Natalia Gutierrez, Benedikt Halldorsson, Okba Hamitou, Guillaume Houzeaux, Stephan Jaure, Mouloud Kessar, Lukas Krenz, Lion Krischer, Soline Laforet, Piero Lanucara, Bo Li, Maria Concetta Lorenzino, Stefano Lorito, Finn Lovholt, Giovanni Macedonio, Jorge Macias, Guillermo Marin, Beatriz Martinez Montesinos, Leonardo Mingari, Genevieve Moguilny, Vadim Montellier, Marisol Monterrubio-Velasco, Georges Emmanuel Moulard, Masaru Nagaso, Massimo Nazaria, Christoph Niethammer, Federica Pardini, Marta Pienkowska, Luca Pizzimenti, Natalia Poiata, Leonhard Rannabauer, Otilio Rojas, Juna Esteban Rodriguez, Fabrizio Romano, Oleksandr Rudyy, Vittorio Ruggiero, Philipp Samfass, Carlos Sanchez-Linares, Sabrina Sanchez, Laura Sandri, Antonio Scala, Nathanael Schaeffer, Joseph Schuchart, Jacopo Selva, Amadine Sergeant, Angela Stallone, Matteo Taroni, Solvi Thrastarson, Manuel Titos, Nadia Tonelllo, Roberto Tonini, Thomas Ulrich, Jean-Pierre Vilotte, Malte Voge, Manuela Volpe, Sara Aniko Wirp, Uwe Woessner

Summary: The EU Center of Excellence for Exascale in Solid Earth (ChEESE) focuses on developing high-level transition capabilities in the geophysics domain to address computational challenges related to earth sciences. The project has successfully optimized community codes and implemented pilot demonstrators to tackle capability and capacity challenges in areas such as seismology, tsunami science, volcanology, and magnetohydrodynamics. The partnership with the Industry and User Board (IUB) has facilitated the dissemination of results across various sectors and institutions.

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE (2023)

Add to Collection

Review Nanoscience & Nanotechnology

Roadmap on quantum nanotechnologies

Arne Laucht, Frank Hohls, Niels Ubbelohde, M. Fernando Gonzalez-Zalba, David J. Reilly, Soren Stobbe, Tim Schroeder, Pasquale Scarlino, Jonne Koski, Andrew Dzurak, Chih-Hwan Yang, Jun Yoneda, Ferdinand Kuemmeth, Hendrik Bluhm, Jarryd Pla, Charles Hill, Joe Salfi, Akira Oiwa, Juha T. Muhonen, Ewold Verhagen, M. D. LaHaye, Hyun Ho Kim, Adam W. Tsen, Dimitrie Culcer, Attila Geresdi, Jan A. Mol, Varun Mohan, Prashant K. Jain, Jonathan Baugh

Summary: Nanostructures have enabled the exploration of quantum behavior, unconventional states, and various quantum phenomena in systems such as nano-optomechanical systems, low-dimensional structures, molecular devices, and more. The intersection of nanotechnology and quantum science/technology has led to significant advancements in quantum research, with applications in quantum metrology, sensing, computing, and communication. This mutual interaction between the two fields has resulted in exciting scientific breakthroughs in the past decade, with promising prospects for the future.

NANOTECHNOLOGY (2021)

Add to Collection

Article Construction & Building Technology

Generation of weather data for the assessment of building performances under future heatwave conditions

Adrien Toesca, Damien David, Kevyn Johannes, Michel Lussault

Summary: This paper presents a methodology for producing a minimum set of heatwave weather files that can represent the diversity of expected heatwaves in a specific location. The methodology involves collecting weather projections, identifying heatwaves, constructing a list of indicators, and selecting representative samples of heatwaves. The methodology was tested for Lyon-Saint Exupery Airport, resulting in the identification of 2229 heatwaves within 1260 years of weather projections. The reduced set of 8 representative heatwave weather files can be used for assessing building performance.

BUILDING AND ENVIRONMENT (2023)

Add to Collection

Article Multidisciplinary Sciences

Anthropogenic influence on extreme temperature and precipitation in Central Asia

Bijan Fallah, Emmanuele Russo, Christoph Menz, Peter Hoffmann, Iulii Didovets, Fred F. Hattermann

Summary: We investigate the contribution of anthropogenic forcing to extreme temperature and precipitation events in Central Asia. Our analysis shows a higher risk of extreme heat events and precipitation events in the region, particularly in Kyrgyzstan and Tajikistan.

SCIENTIFIC REPORTS (2023)

Add to Collection

Article Meteorology & Atmospheric Sciences

The Impact of an Extreme Cloud burst on Edinburgh Castle

Simon F. B. Tett, Young Hwa Cha, Kate Donovan, Gina-Maria Geffers, Ed Hawkins

Summary: This passage indicates that an extreme cloudburst on 4 July 2021 caused damage to Edinburgh Castle, and based on the results from a convective permitting model, the observed warming increased the risk by approximately 30% compared to preindustrial levels, which further increases to about 70% in a +2°C world.

BULLETIN OF THE AMERICAN METEOROLOGICAL SOCIETY (2023)

Add to Collection

Article Engineering, Civil

EQSIM-A multidisciplinary framework for fault-to-structure earthquake simulations on exascale computers, part II: Regional simulations of building response

David McCallen, Floriana Petrone, Mamun Miah, Arben Pitarka, Arthur Rodgers, Norman Abrahamson

Summary: The current observational database on regional strong ground motions and building response for major earthquakes is limited, but emerging high-resolution, physics-based ground motion simulations offer new opportunities for studying the distribution of ground motions and infrastructure demands.

EARTHQUAKE SPECTRA (2021)

Add to Collection

Article Computer Science, Information Systems

Scalability and efficiency challenges for the exascale supercomputing system: practice of a parallel supporting environment on the Sunway exascale prototype system

Xiaobin He, Xin Chen, Heng Guo, Xin Liu, Dexun Chen, Yuling Yang, Jie Gao, Yunlong Feng, Longde Chen, Xiaona Diao, Zuoning Chen

Summary: With the advancement of supercomputer performance and the integration of artificial intelligence with traditional scientific computing, parallel applications have scaled from millions to tens of millions of computing cores. This poses challenges in achieving high scalability and efficiency on super-large-scale systems. This paper analyzes the challenges faced by parallel applications in the exascale era using the Sunway exascale prototype system as an example. It highlights optimization technologies employed in the parallel supporting environment software, such as the parallel operating system, I/O optimization technology, debugging technology, parallel algorithm, and mixed-precision method. The contributions to various applications running on the Sunway exascale prototype system are also discussed, showcasing the effectiveness of the parallel supporting environment design.

FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING (2023)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Projecting LBM performance on Exascale class Architectures: A tentative outlook

Giorgio Amati, Sauro Succi, Pierluigi Fanelli, Vesselin K. Krastev, Giacomo Falcucci

Summary: The paper presents a prospective scenario for using a Lattice Boltzmann (LB) code on forthcoming Exascale supercomputers based on current Petascale HPC performance, aiming to provide guidelines for planning larger and more complex simulations in the future.

JOURNAL OF COMPUTATIONAL SCIENCE (2021)

Add to Collection

Article Engineering, Marine

Reflecting the uncertainties of ensemble weather forecasts on the predictions of ship fuel consumption

R. Vettor, C. Guedes Soares

Summary: This study investigates different methodologies for predicting the uncertainties in ship fuel consumption, including ensemble weather forecasts, probabilistic approaches, and a classical first-order second-moment method. By comparing the results from different methods, it is found that the actual fuel consumption falls within the predicted range with reasonable agreement.

OCEAN ENGINEERING (2022)

Add to Collection

Article Geosciences, Multidisciplinary

Advancing precipitation prediction using a new-generation storm-resolving model framework - SIMA-MPAS (V1.0): a case study over the western United States

Xingying Huang, Andrew Gettelman, William C. Skamarock, Peter Hjort Lauritzen, Miles Curry, Adam Herrington, John T. Truesdale, Michael Duda

Summary: This study uses a state-of-the-art storm-resolving global climate model (SIMA-MPAS) to simulate storms and precipitation, showing more realistic regional climate variability and fine-scale features. Comparisons with traditional regional climate models demonstrate the advantages of SIMA-MPAS.

GEOSCIENTIFIC MODEL DEVELOPMENT (2022)

Add to Collection

Article Nanoscience & Nanotechnology

Roadmap on emerging hardware and technology for machine learning

Karl Berggren, Qiangfei Xia, Konstantin K. Likharev, Dmitri B. Strukov, Hao Jiang, Thomas Mikolajick, Damien Querlioz, Martin Salinga, John R. Erickson, Shuang Pi, Feng Xiong, Peng Lin, Can Li, Yu Chen, Shisheng Xiong, Brian D. Hoskins, Matthew W. Daniels, Advait Madhavan, James A. Liddle, Jabez J. McClelland, Yuchao Yang, Jennifer Rupp, Stephen S. Nonnenmann, Kwang-Ting Cheng, Nanbo Gong, Miguel Angel Lastras-Montano, A. Alec Talin, Alberto Salleo, Bhavin J. Shastri, Thomas Ferreira de Lima, Paul Prucnal, Alexander N. Tait, Yichen Shen, Huaiyu Meng, Charles Roques-Carmes, Zengguang Cheng, Harish Bhaskaran, Deep Jariwala, Han Wang, Jeffrey M. Shainline, Kenneth Segall, J. Joshua Yang, Kaushik Roy, Suman Datta, Arijit Raychowdhury

Summary: Recent progress in artificial intelligence is primarily attributed to the rapid development of machine learning, but the performance and energy efficiency of hardware systems set fundamental limits on machine learning capabilities. Data-centric computing requires a revolution in hardware systems, with new hardware platforms offering hope for future computing with improved throughput and energy efficiency. However, challenges such as materials selection, device optimization, circuit fabrication, and system integration must be addressed in building such systems.

NANOTECHNOLOGY (2021)

Add to Collection

Article Environmental Sciences

On the attribution of the impacts of extreme weather events to anthropogenic climate change

S. E. Perkins-Kirkpatrick, D. A. Stone, D. M. Mitchell, S. Rosier, A. D. King, Y. T. E. Lo, J. Pastor-Paz, D. Frame, M. Wehner

Summary: Investigations into the role of anthropogenic climate change in extreme weather events need to carefully consider methodological choices to avoid misinterpretation. The anthropogenic signal behind weather events may differ from the signal behind the impact system, and it is important to assess impacts in addition to meteorological events due to lags and nonlinearities.

ENVIRONMENTAL RESEARCH LETTERS (2022)

Add to Collection

Correction Meteorology & Atmospheric Sciences

The Impact of Three-Dimensional Effects on the Simulation of Turbulence Kinetic Energy in a Major Alpine Valley (vol 168, pg 1, 2018)

Brigitta Goger, Mathias W. Rotach, Alexander Gohm, Oliver Fuhrer, Ivana Stiperski, Albert A. M. Holtslag

BOUNDARY-LAYER METEOROLOGY (2023)

Add to Collection

Article Environmental Sciences

Climate Scenarios for Switzerland CH2018-Approach and Implications

A. M. Fischer, K. M. Strassmann, M. Croci-Maspoli, A. M. Hama, R. Knutti, S. Kotlarski, C. Schar, C. Schnadt Poberaj, N. Ban, M. Bavay, U. Beyerle, D. N. Bresch, S. Bronnimann, P. Burlando, A. Casanueva, S. Fatichi, I Feigenwinter, E. M. Fischer, M. Hirschi, M. A. Liniger, C. Marty, I Medhaug, N. Peleg, M. Pickl, C. C. Raible, J. Rajczak, O. Rossler, S. C. Scherrer, C. Schwierz, S. Seneviratne, M. Skelton, S. L. Sorland, C. Spirig, F. Tschurr, J. Zeder, E. M. Zubler

Summary: To address climate change, Switzerland has developed user-oriented climate change scenarios through close collaboration between academia and administration. These scenarios show that Switzerland will face dry summers, heavy precipitation, more hot days, and snow-scarce winters in the future. Approximately half of these changes could be alleviated by strong global mitigation efforts by the mid-century.

CLIMATE SERVICES (2022)

Add to Collection

Article Computer Science, Hardware & Architecture

The Convergence of Hyperscale Data Center and High-Performance Computing Networks

Torsten Hoefler, Ariel Hondo, Duncan Roweth

Summary: This article discusses the differences and similarities between network technologies used in supercomputers and data centers, and outlines a path to convergence at multiple layers. It predicts that emerging smart networking solutions will accelerate this convergence.

COMPUTER (2022)

Add to Collection

Article Computer Science, Hardware & Architecture

Benchmarking Data Science: 12 Ways to Lie With Statistics and Performance on Parallel Computers

Torsten Hoefler

Summary: This article humorously discusses 12 fallacies in compute performance that are frequently observed in practice, providing recommendations to mitigate the associated dangers and contributing to good benchmarking etiquette for data science.

COMPUTER (2022)

Add to Collection

Article Meteorology & Atmospheric Sciences

Aerosol sensitivity simulations over East Asia in a convection-permitting climate model

Shuping Li, Silje Lund Sorland, Martin Wild, Christoph Schaer

Summary: The parameterization of deep convection is a major source of uncertainties in regional climate simulations. Due to limitations in computational resources, long-term simulations at kilometer-scale resolution with explicit deep convection have been limited, particularly over East Asia. To address this, a pair of 10-year reference simulations were conducted, one using convection parameterization and 12 km resolution, and the other using convection-permitting simulation at 4.4 km resolution.

CLIMATE DYNAMICS (2023)

Add to Collection

Article Meteorology & Atmospheric Sciences

Systematic Calibration of a Convection-Resolving Model: Application Over Tropical Atlantic

Shuchang Liu, Christian Zeman, Silje Lund Sorland, Christoph Schar

Summary: Calibrating non-hydrostatic km-scale weather and climate models leads to significant improvements in simulating clouds and precipitation. Systematic calibration improves model performance, benefiting model intercomparison projects, process studies, and climate-change scenario simulations.

JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES (2022)

Add to Collection

Review Computer Science, Hardware & Architecture

Disentangling Hype from Practicality: On Realistically Achieving Quantum Advantage

Torsten Hoefler, Thomas Haener, Matthias Troyer

Summary: Operating on different principles, quantum computers can solve important problems that were previously unsolvable on classical computers. The time to solve certain problems on a quantum computer grows more slowly with the size of the problem, which is called quantum speedup. Beyond quantum supremacy, the focus is on finding meaningful applications that can be realistically solved faster on a quantum computer than on a classical one, known as quantum practicality.

COMMUNICATIONS OF THE ACM (2023)

Add to Collection

Article Meteorology & Atmospheric Sciences

Simulating Hail and Lightning Over the Alpine Adriatic Region-A Model Intercomparison Study

B. Malecic, R. Cui, M. E. Demory, K. Horvath, D. Jelic, C. Schar, M. Telisman Prtenjak, P. Velasquez, N. Ban

Summary: Hail is a significant convective weather hazard, and it is challenging to model and forecast due to limited computational resolution and gaps in understanding hail formation. This study analyzes eight hailstorms using simulations and model intercomparison to assess the ability of different modeling systems in reproducing convective extremes. The results indicate that both HAILCAST and LPI are valuable tools for real-time forecasting and climatological assessment of hail and lightning in current and changing climate.

JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES (2023)

Add to Collection

Article Meteorology & Atmospheric Sciences

Application of the Pseudo-Global Warming Approach in a Kilometer-Resolution Climate Simulation of the Tropics

Christoph Heim, David Leutwyler, Christoph Schar

Summary: Clouds over tropical oceans play a crucial role in the Earth's response to greenhouse gas concentrations, but modeling their behavior is challenging due to small-scale convective processes. In this study, a kilometer-resolution simulation (COSMO) is compared with global climate models (GCMs) to understand their response to global warming. The COSMO simulation shows a different behavior, with no double ITCZ bias, making it valuable for further research.

JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES (2023)

Add to Collection

Article Geosciences, Multidisciplinary

How well does a convection-permitting regional climate model represent the reverse orographic effect of extreme hourly precipitation?

Eleonora Dallan, Francesco Marra, Giorgia Fosser, Marco Marani, Giuseppe Formetta, Christoph Schar, Marco Borga

Summary: High-resolution convection-permitting models (CPMs) are crucial for estimating future short-duration extreme precipitation in mountainous regions. However, recent observational studies indicate that extreme hourly precipitation tends to decrease with elevation, which may be related to subgrid processes not fully simulated by CPMs. To evaluate the reliability of CPM projections, it is important to understand their ability to capture this reverse orographic effect.

HYDROLOGY AND EARTH SYSTEM SCIENCES (2023)

Add to Collection

Article Geosciences, Multidisciplinary

The pseudo-global-warming (PGW) approach: methodology, software package PGW4ERA5 v1.1, validation, and sensitivity analyses

Roman Brogli, Christoph Heim, Jonas Mensch, Silje Lund Sorland, Christoph Schar

Summary: The term pseudo-global warming (PGW) refers to a simulation strategy in regional climate modeling where large-scale changes in the climate system are directly imposed on a control regional climate simulation by modifying the boundary conditions. This approach offers benefits such as lower computational requirements and flexibility in simulation design. However, care must be taken not to compromise the physics of the regional climate model when implementing PGW simulations. To simplify the preparation of PGW simulations, a detailed methodology and companion software are provided.

GEOSCIENTIFIC MODEL DEVELOPMENT (2023)

Add to Collection

Proceedings Paper Computer Science, Hardware & Architecture

SAGE: Software-based Attestation for GPU Execution

Andrei Ivanov, Benjamin Rothenberger, Arnaud Dethise, Marco Canini, Torsten Hoefler, Adrian Perrig

Summary: With the increasing application of machine learning in security-critical domains, ensuring integrity and privacy in computation using accelerators like GPUs has become a growing concern. This study investigates the limited support for trusted execution on GPUs and presents a purely software-based approach called SAGE for secure GPU code execution. The evaluation shows that SAGE is practical for achieving trustworthy execution on GPUs even without specific hardware support.

PROCEEDINGS OF THE 2023 USENIX ANNUAL TECHNICAL CONFERENCE (2023)

Add to Collection

Proceedings Paper Computer Science, Hardware & Architecture

Productive Performance Engineering for Weather and Climate Modeling with Python

Tal Ben-Nun, Linus Groner, Florian Deconinck, Tobias Wicky, Eddie Davis, Johann Dahm, Oliver D. Elbert, Rhea George, Jeremy McGibbon, Lukas Trumper, Elynn Wu, Oliver Fuhrer, Thomas Schulthess, Torsten Hoefler

Summary: This study presents an optimization approach for weather and climate applications, achieving significant performance improvements by abstracting hardware details and utilizing optimization techniques.

SC22: INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS (2022)

Add to Collection

Proceedings Paper Computer Science, Hardware & Architecture

ProbGraph: High-Performance and High-Accuracy Graph Mining with Probabilistic Set Representations

Maciej Besta, Cesare Miglioli, Paolo Sylos Labini, Jakub Tetek, Patrick Iff, Raghavendra Kanakagiri, Saleh Ashkboos, Kacper Janda, Michal Podstawski, Grzegorz Kwasniewski, Niels Gleinig, Flavio Vella, Onur Mutlu, Torsten Hoefler

Summary: This paper proposes ProbGraph, a graph representation that accelerates graph mining problems with strong theoretical guarantees on work, depth, and result accuracy. By using probabilistic set representations such as Bloom filters, the algorithm achieves significant speedup while ensuring accuracy. The novel bounds and algorithms based on probabilistic set representations have separate interest for the data analytics community.

SC22: INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS (2022)

Add to Collection

Proceedings Paper Computer Science, Hardware & Architecture

Building Blocks for Network-Accelerated Distributed File Systems

Salvatore Di Girolamo, Daniele De Sensi, Konstantin Taranov, Milos Malesevic, Maciej Besta, Timo Schneider, Severin Kistler, Torsten Hoefler

Summary: Storage systems in high-performance clusters and datacenters require improved performance. Remote direct memory access (RDMA) and SmartNICs have been used to accelerate data flow and offload storage policies, resulting in improved latency and CPU utilization.

SC22: INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS (2022)

Add to Collection

No Data Available

© Peeref 2019-2024. All rights reserved.