Filters








6,827 Hits in 7.5 sec

Software Abstractions and Methodologies for HPC Simulation Codes on Future Architectures

2014 Journal of Open Research Software  
Large, complex, multi-scale, multi-physics simulation codes, running on high performance com-puting (HPC) platforms, have become essential to advancing science and engineering.  ...  These codes simulate multi-scale, multi-physics phenomena with unprecedented fidelity on petascale platforms, and are used by large communities.  ...  " highlevel abstraction. • Performance Portability.  ... 
doi:10.5334/jors.aw fatcat:drphessgufbxph3hrwngypvkd4

Preface: Summit and Sierra Supercomputers

Jaime H. Moreno, Yoonho Park
2020 IBM Journal of Research and Development  
implemented using RAJA and CUDA, and estimates the performance penalty arising from the portability abstraction layer.  ...  The fourth article, "Building a High-Performance Resilient Scalable Storage Cluster for CORAL Using IBM ESS" by Islam and Shah, describes the design and implementation of the high-performance, scalable  ... 
doi:10.1147/jrd.2020.2976169 fatcat:f7r5b2awubej5lribdge4e64ea

Optimizations to enhance sustainability of MPI applications

Jesus Carretero, Horacio Perez-Sanchez, Jose M. Cecilia, Javier Garcia-Blas, David E. Singh, Florin Isaila, Thomas Fahringer, Radu Prodan, George Bosilca, Alexey Lastovetsky, Christi Symeonidou
2014 Proceedings of the 21st European MPI Users' Group Meeting on - EuroMPI/ASIA '14  
In this paper, we layout some proposals to extend MPI to cover major relevant domains in a move towards sustainability, including: MPI programming optimizations and programming models, resilience, data  ...  However, to achieve this level of performance, we need to design and implement more sustainable solutions for ultra-scale computing systems, at both the hardware and software levels, while understanding  ...  Aside from applications, these new interfaces can be used by high-level abstractions, such as transactional fault tolerance, uncoordinated checkpointrestart, and programming languages, to support advanced  ... 
doi:10.1145/2642769.2642797 dblp:conf/pvm/CarreteroGSIFPBLSPC14 fatcat:dpbi7wueovg7jhyoy6chxk43l4

Fault-tolerant communication runtime support for data-centric programming models

Abhinav Vishnu, Huub Van Dam, Wibe De Jong, Pavan Balaji, Shuaiwen Song
2010 2010 International Conference on High Performance Computing  
In this paper, we take a first step toward data-centric fault resilience by designing and implementing a fault0resilient, onesided communication runtime framework using Global Arrays and its communication  ...  Our performance evaluation indicates that the framework incurs little overhead compared to state-of-the-art designs and provides a fundamental framework of fault resiliency for PGAS models.  ...  Fault-Resilient Non-Data-Moving Collective Communication Primitives Collective communication primitives are widely used by programming models to provide abstractions for processes in a group to perform  ... 
doi:10.1109/hipc.2010.5713195 dblp:conf/hipc/VishnuDJBS10 fatcat:lt2cmbbk3vdtjggztbx22njcca

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  ...  The Landscape of Exascale Research: A Data-Driven Literature Analysis 23:27 Toward Exascale Resilience Over the past few years resilience has became a major issue for high-performance computing (HPC)  ... 
doi:10.1145/3372390 fatcat:jhtwt7pxd5c5darhz75hiqgsnq

Navigating the Seismic Shift of Post-Moore Computer Systems Design

Anindya Banerjee, Sankar Basu, Erik Brunvand, Pinaki Mazumder, Walter R. Cleaveland II, Gurdip Singh, Margaret Martonosi, Fernanda Pembleton
2021 IEEE Micro  
and high-performance, and secure computer systems from the underlying hardware that is complex and heterogeneous.  ...  PPoSS expects coordinated progress at the intersection of multiple disciplines including, but not limited to, computer architecture, high-performance computing, programming languages and compilers, machine  ... 
doi:10.1109/mm.2021.3114899 fatcat:5ukjfbfawnhb5iangpskky5n4e

Lyncs-API: a Python API for Lattice QCD applications [article]

Simone Bacchio, Jacob Finkenrath, Christodoulos Stylianou
2022 arXiv   pre-print
the same high-level workflow.  ...  Lyncs flexibly links to libraries for CPUs and GPUs in a way that can accommodate additional computing architectures as these arise, achieving performance-portability for the calculations while maintaining  ...  Introduction Three keywords in High-Performance Computing (HPC) are Performance, Portability and Productivity, known as the three "P".  ... 
arXiv:2201.03873v1 fatcat:xz2i3g5etbdclh4bhpkwj6pk6y

Acceleration of MPI mechanisms for sustainable HPC applications

2015 Supercomputing Frontiers and Innovations  
In this paper, we show some proposals to extend MPI trying to cover major domains that are relevant towards sustainability: MPI programming optimizations and programming models, resilience, data management  ...  In this paper, we layout some proposals to extend MPI to cover major relevant domains in a move towards sustainability, including: MPI programming optimizations and programming models, resilience, data  ...  Aside from applications, these new interfaces can be used by high-level abstractions, such as FMI, FTI, CS, transactional fault tolerance, uncoordinated checkpoint-restart, and programming languages, to  ... 
doi:10.14529/jsfi150202 fatcat:hnu3cj5nwzhmjccfwa2drudck4

A Personal, Historical Perspective of Parallel Programming for High Performance [chapter]

Christian Lengauer
2000 Communication-Based Systems  
To me, the most dominant issue in high-performance computing, which has hardly been addressed effectively so far, is that of portable performance.  ...  There are other aspects of Java which make it seem like an unlikely candidate for high-performance computing: its thread model is inefficient, the array -the data structure used most for high performance-cannot  ... 
doi:10.1007/978-94-015-9608-4_9 fatcat:tpc33ohn3fgapcfrmut7fiya5i

Reliable heterogeneous applications

J. Lee, S.J. Chapin, S. Taylor
2003 IEEE Transactions on Reliability  
Computational resiliency involves the dynamic use of replicated software structures, guided by mission policy, to achieve reliable operation.  ...  Heterogeneous load balancing of replicated structures for performance improvement. 4. Demonstration of technologies using typical real-world applications in various fields. 5.  ...  Computational Resiliency Using resiliency, periodic liveness checks are performed.  ... 
doi:10.1109/tr.2003.819502 fatcat:bccf3vwx7fhfpfqzd6hcp6tbai

IPDPS 2021 PhD Forum Welcome and Abstracts

2021 2021 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)  
Description The IEEE Computer Society has created the Technical Consortium on High Performance Computing (TCHPC) to advance and coordinate work in the field of high performance computing networking, storage  ...  It has grown from the traditional poster presentations by students working toward a PhD in broadly defined areas related to parallel and distributed processing to a broader, enhanced program to include  ...  The task performance is modeled using an α-β-model. Preliminary results show high accuracy for compute kernels.  ... 
doi:10.1109/ipdpsw52791.2021.00160 fatcat:c5srphikvba6pklpxsyx6lfmhy

The Omni Macroprogramming Environment for Sensor Networks [chapter]

Asad Awan, Ahmed Sameh, Ananth Grama
2006 Lecture Notes in Computer Science  
In this paper, we provide a high-level overview of the Omni architecture, its salient features, and implementation details.  ...  The Omni architecture is designed to be a flexible, extensible, scalable, and portable system, upon which a wide variety of DDDAS applications can be built.  ...  The Omni architecture provides a flexible, portable, and scalable platform over which a number of DDDAS applications can be built.  ... 
doi:10.1007/11758532_62 fatcat:qcbqkhwg2ngolmmtesngchcuii

Paving the Way Towards a Highly Energy-Efficient and Highly Integrated Compute Node for the Exascale Revolution: The ExaNoDe Approach

Alvise Rigo, Christian Pinto, Kevin Pouget, Daniel Raho, Denis Dutoit, Pierre-Yves Martinez, Chris Doran, Luca Benini, Iakovos Mavroidis, Manolis Marazakis, Valeria Bartsch, Guy Lonsdale (+8 others)
2017 2017 Euromicro Conference on Digital System Design (DSD)  
Power consumption and high compute density are the key factors to be considered when building a compute node for the upcoming Exascale revolution.  ...  research project aiming to design a highly energy efficient and highly integrated heterogeneous compute node targeting Exascale level computing, mixing low-power processors, heterogeneous co-processors and using  ...  The work presented in this paper reflects only authors' view and the European Commission is not responsible for any use that may be made of the information it contains.  ... 
doi:10.1109/dsd.2017.37 dblp:conf/dsd/RigoPPRDMDBMMBL17 fatcat:jumekx7n6vcmvnu32epz3tp6py

IMPROVING PERFORMANCE IN HPC SYSTEM UNDER POWER CONSUMPTIONS LIMITATIONS

Muhammad Usman Ashraf
2019 International Journal of Advanced Research in Computer Science  
towards everyday life.  ...  Current supercomputers can't achieve such a high level of performance under power dissipation constraints.  ...  Resiliency There is a need for considerably new computing strategies for having the roadmap towards Exascale computing environment.  ... 
doi:10.26483/ijarcs.v10i2.6397 fatcat:k3l3lk5kuzhnldn5b2qzkh4eia

Empirical Comparison of Three Versioning Architectures

Hajime Fujita, Kamil Iskra, Pavan Balaji, Andrew A. Chien
2015 2015 IEEE International Conference on Cluster Computing  
We use a synthetic workload comparing the three array architectures in terms of runtime performance and memory requirements.  ...  We are developing the Global View Resilience (GVR) library, with multi-version global arrays as one of the key features.  ...  , portable, and flexible error handling.  ... 
doi:10.1109/cluster.2015.69 dblp:conf/cluster/FujitaIBC15 fatcat:zf75njpuqbgjlehx4pllag3bji
« Previous Showing results 1 — 15 out of 6,827 results