617 Hits in 4.8 sec

Towards Portable Online Prediction of Network Utilization Using MPI-Level Monitoring [chapter]

Shu-Mei Tseng, Bogdan Nicolae, George Bosilca, Emmanuel Jeannot, Aparna Chandramowlishwaran, Franck Cappello
2019 Lecture Notes in Computer Science  
Related Work MPI monitoring. There are many different ways to monitor the network utilization of an MPI application.  ...  The key novelty of our approach is two-fold: (1) we devise a mechanism to approximate network utilization using only the information available at the MPI-level (which addresses the portability challenge  ...  Portable MPI-level monitoring. To solve the problem of portable monitoring, we propose to capture the network utilization directly from the communication library, in this context MPI.  ... 
doi:10.1007/978-3-030-29400-7_4 fatcat:f6hqpqep5bfl7ckqrhk4c7jcse

A Transition from Traditional Checkpointing towards Multi-Agent based Approaches

Gerard McKee, Blesson Varghese, Vassil Alexandrov
2010 Journal of clean energy technologies  
The approach is implemented on a computer cluster and experimental results are presented to validate the feasibility of the approach and its contribution towards enhancing fault tolerance.  ...  Five resources namely, the executed problem, parallel computing platform, middleware, hardware abstraction and agents that contribute towards the infrastructure of the proposed approach is considered.  ...  Thirdly, an agent can predict core failures by consistent monitoring (for example, power consumption and heat dissipation of the cores can be used to predict failures).  ... 
doi:10.7763/ijcte.2010.v2.228 fatcat:fryttfct2vhmjon7xwpobilmni

Energy-efficient Algorithms for Ultrascale Systems

2015 Supercomputing Frontiers and Innovations  
In the software area, there also exist a multitude of research approaches towards energy saving, often concentrating either on the system software level or the application organization level, reflecting  ...  With the world's most powerful systems already well past the Petaflop/s level in 2014, a projection of this trend leads to the prediction that by 2022, Exascale computing will be possible.  ...  For CPU power monitoring, one approach consists in finding the relationship between the power consumption and the utilization level.  ... 
doi:10.14529/jsfi150205 fatcat:hceabokapvgozc4tsgikcjyq3u


Fang Zheng, Hongfeng Yu, Can Hantas, Matthew Wolf, Greg Eisenhauer, Karsten Schwan, Hasan Abbasi, Scott Klasky
2013 Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '13  
This involves recognizing the potential causes of on-node resource contention and then using scheduling methods that prevent them.  ...  GoldRush uses fine-grained scheduling to "steal" idle resources, in ways that minimize interference between the simulation and in situ analytics.  ...  Department of Energy, and Center for Exascale Simulation of Combustion in Turbulence, U.S. Department of Energy.  ... 
doi:10.1145/2503210.2503279 dblp:conf/sc/ZhengYHWESAK13 fatcat:4wfqrtk7kfbuvbxrovhrgvn5my

Tools for Analyzing Parallel I/O [chapter]

Julian Martin Kunkel, Eugen Betke, Matt Bryson, Philip Carns, Rosemary Francis, Wolfgang Frings, Roland Laifer, Sandra Mendez
2018 Lecture Notes in Computer Science  
In this paper, we highlight advances in monitoring tools to help address these issues.  ...  Department of Energy, Office of Science, under contract DE-AC02-06CH11357. We thank Felicity Pitchers for the proofreading.  ...  Above the parallel file systems are MPI-IO, MPI libraries, and upper-level libraries such as HDF5.  ... 
doi:10.1007/978-3-030-02465-9_4 fatcat:heg3otk57jd3dlfkmukjh5kpfq

Tools for Analyzing Parallel I/O [article]

Julian M. Kunkel, Eugen Betke, Matt Bryson, Philip Carns, Rosemary Francis, Wolfgang Frings, Roland Laifer, Sandra Mendez
2018 arXiv   pre-print
In this paper, we highlight advances in monitoring tools to help address this problem.  ...  Department of Energy, Office of Science, under contract DE-AC02-06CH11357. We thank Felicity Pitchers for the proofreading.  ...  Above the parallel file systems are MPI-IO, MPI libraries, and upper-level libraries such as HDF5.  ... 
arXiv:1807.04985v3 fatcat:my7aswjrhjfi3fvx6m6y76jbya

Integrating hydrological modelling, data assimilation and cloud computing for real-time management of water resources

Wolfgang Kurtz, Andrei Lapin, Oliver S. Schilling, Qi Tang, Eryk Schiller, Torsten Braun, Daniel Hunkeler, Harry Vereecken, Edward Sudicky, Peter Kropf, Harrie-Jan Hendricks Franssen, Philip Brunner
2017 Environmental Modelling & Software  
The changes in the model codes for the utilization of the cloud infrastructure were minimal and mainly concerned the forward propagation of the model ensemble.  ...  A synthetic data assimilation experiment based on the widely used tilted V-catchment problem served as a benchmark for testing the usefulness of the proposed modelling platform and for evaluating the computational  ...  for hydrogeological use-cases.  ... 
doi:10.1016/j.envsoft.2017.03.011 fatcat:cfhhhxzso5bgxavb5jaeclxjnm

Data Mining in Ubiquitous Healthcare [chapter]

Viswanathan, Whangbo, Yang
2011 New Fundamental Technologies in Data Mining  
If the number of clusters exceeds the available number of hosts then some hosts will classify multiple clusters (using MPI) .  ...  There are several levels of data mining involved in this process.  ...  The series of books entitled by "Data Mining" address the need by presenting in-depth description of novel mining algorithms and many useful applications.  ... 
doi:10.5772/14390 fatcat:modnn7ckmbcozcbmbhbssviwlu

On Processing Extreme Data

Dana Petcu, Gabriel Iuhasz, Daniel Pop, Domenico Talia, Jesus Carretero, Radu Prodan, Thomas Fahringer, Ivan Grasso, Ramon Doallo, Maria J. Martin, Basilio B. Fraguela, Roman Trobec (+11 others)
2016 Scalable Computing : Practice and Experience  
It has been found that the majority of applications prefer to use POSIX interfaces rather than the MPI-IO interface or high-level libraries.  ...  Velocity dimension using new interconnection network topologies).  ... 
doi:10.12694/scpe.v16i4.1134 fatcat:yibmtpz5szgudgocgoojodhjc4

Linux and Supercomputing: How My Passion for Building COTS Systems Led to an HPC Revolution

David A. Bader, Alex Magoun
2021 IEEE Annals of the History of Computing  
For MPI, Roadrunner used MPICH, a high performance open-source MPI implementation from Argonne National Laboratory; Myricom GM network drivers; and MPICH GM, Myricom's MPI implementation.  ...  The Myrinet System Area Network (Myrinet/SAN) interconnection network was one of Roadrunner was among the 100 fastest supercomputers in the world when it went online.  ... 
doi:10.1109/mahc.2021.3101415 fatcat:mbbyfho3sbfzjha4377piwkpya

Tackling Exascale Software Challenges in Molecular Dynamics Simulations with GROMACS [chapter]

Szilárd Páll, Mark James Abraham, Carsten Kutzner, Berk Hess, Erik Lindahl
2015 Lecture Notes in Computer Science  
Release 4.6 of GROMACS uses SIMD acceleration on a wide range of architectures, GPU offloading acceleration, and both OpenMP and MPI parallelism within and between nodes, respectively.  ...  Here, we describe some of the ways we have been able to realize this through the use of parallelization on all levels, combined with a constant focus on absolute performance.  ...  GROMACS uses the Redmine issue-tracking system 3 to discuss feature development, report and discuss bugs, and to monitor intended and actual progress towards milestones.  ... 
doi:10.1007/978-3-319-15976-8_1 fatcat:rnekhgugmfe6zgpgwsybw4ywey

Energy-Aware High-Performance Computing: Survey of State-of-the-Art Tools, Techniques, and Environments

Pawel Czarnul, Jerzy Proficz, Adam Krzywaniak
2019 Scientific Programming  
., applications and benchmarks used in particular works are discussed.  ...  We discuss tools and APIs for energy/power management as well as tools and environments for prediction and/or simulation of energy/power consumption in modern HPC systems.  ...  Topics related to power monitoring for ultrascale systems are presented in [17] . e paper describes solutions used for online power measurement, including a profound analysis of the current state-of-the-art  ... 
doi:10.1155/2019/8348791 fatcat:ib3dvjzg2bhhjnnklb4kaj2eqi

ExaWorks: Workflows for Exascale [article]

Aymen Al-Saadi, Dong H. Ahn, Yadu Babuji, Kyle Chard, James Corbett, Mihael Hategan, Stephen Herbein, Shantenu Jha, Daniel Laney, Andre Merzky, Todd Munson, Michael Salim (+3 others)
2021 arXiv   pre-print
We present the ExaWorks project, which can address many of these challenges: ExaWorks is leading a co-design process to create a workflow software development Toolkit (SDK) consisting of a wide range of  ...  These software combinations and integrations, however, are difficult to achieve due to challenges of coordination and deployment of heterogeneous software components on diverse and massive platforms.  ...  We thank Rafael Ferreira da Silva and Henri Casanova of the Workflows-RI project for partnering to organize community summits.  ... 
arXiv:2108.13521v1 fatcat:y4qe4oxscbfhhkgydqz7lft3cq

The Landscape of Exascale Research

Stijn Heldens, Pieter Hijma, Ben Van Werkhoven, Jason Maassen, Adam S. Z. Belloum, Rob V. Van Nieuwpoort
2020 ACM Computing Surveys  
We use a three-stage approach in which we (1) discuss various exascale landmark studies, (2) use data-driven techniques to analyze the large collection of related literature, and (3) discuss eight research  ...  However, as we look forward, we still foresee two major concerns: the lack of suitable programming tools and the growing gap between processor performance and data bandwidth (i.e., memory, storage, networks  ...  Programming is still often done using low-level solutions, such as Fortran and MPI.  ... 
doi:10.1145/3372390 fatcat:jhtwt7pxd5c5darhz75hiqgsnq

Cloud Computing and Big Data: A Review of Current Service Models and Hardware Perspectives

Richard Branch, Heather Tjeerdsma, Cody Wilson, Richard Hurley, Sabine McConnell
2014 Journal of Software Engineering and Applications  
Big Data applications are pervading more and more aspects of our life, encompassing commercial and scientific uses at increasing rates as we move towards exascale analytics.  ...  most important concepts of cloud computing as it relates to processing of Big Data.  ...  Resource monitoring is leveraged in order to provide transparent resource utilization reporting to the cloud user and provider.  ... 
doi:10.4236/jsea.2014.78063 fatcat:5s62y7h6ozh5nbj4sfpgpfol5u
« Previous Showing results 1 — 15 out of 617 results