Filters








43 Hits in 5.7 sec

Enabling Parallel Simulation of Large-Scale HPC Network Systems

Misbah Mubarak, Christopher D. Carothers, Robert B. Ross, Philip Carns
2017 IEEE Transactions on Parallel and Distributed Systems  
In this work, we present a simulation framework for modeling two important classes of networks used in today's IBM and Cray supercomputers: torus and dragonfly networks.  ...  Most state-of-the-art HPC network simulation frameworks, however, are constrained in one or more of these areas.  ...  We are thankful to Nikhil Jain (UIUC) for his contributions to CODES network models.  ... 
doi:10.1109/tpds.2016.2543725 fatcat:4fmce73wyvdsrlkhfdqrldfhva

Low latency network and distributed storage for next generation HPC systems: the ExaNeSt project

R Ammendola, A Biagioni, P Cretaro, O Frezza, F Lo Cicero, A Lonardo, M Martinelli, P S Paolucci, E Pastorelli, F Pisani, F Simula, P Vicini (+5 others)
2017 Journal of Physics, Conference Series  
The adoption of low-cost, Linux-based clusters extended the reach of HPC from its roots in modelling and simulation of complex physical systems to a broader range of industries, from biotechnology, cloud  ...  , low latency and high performance network and equipped with a distributed storage architecture.  ...  Network simulator results Benchmarking and characterization of an interconnection network depend on many parameters -e.g. traffic pattern, buffers and network sizes -therefore at very large scale simulations  ... 
doi:10.1088/1742-6596/898/8/082045 fatcat:7tytbz7ao5gwxh6oxoxrxcrmxm

On-the-Fly Adaptive Routing in High-Radix Hierarchical Networks

Marina Garcia, Enrique Vallejo, Ramon Beivide, Miguel Odriozola, Cristobal Camarero, Mateo Valero, Germ'n Rodriguez, Jesus Labarta, Cyriel Minkenberg
2012 2012 41st International Conference on Parallel Processing  
Dragonfly networks have been recently proposed for the interconnection network of forthcoming exascale supercomputers.  ...  Relying on large-radix routers, they build a topology with low diameter and high throughput, divided into multiple groups of routers.  ...  ACKNOWLEDGEMENTS This work has been supported by the Spanish Ministerio de Ciencia e Innovación, under project TIN2010-21291-C02-02, The HiPEAC Network of Excelence, the Consolider Project CSD2007-00050  ... 
doi:10.1109/icpp.2012.46 dblp:conf/icpp/GarciaVBOCVRLM12 fatcat:5rchz4vm7rhh7ilfw6by75o7ey

SpectralFly: Ramanujan Graphs as Flexible and Efficient Interconnection Networks [article]

Stephen Young, Sinan Aksoy, Jesun Firoz, Roberto Gioiosa, Tobias Hagge, Mark Kempton, Juan Escobedo, Mark Raugas
2022 arXiv   pre-print
Additionally, we simulate the performance of SpectralFly on a representative sample of micro-benchmarks using the Structure Simulation Toolkit Macroscale Element Library simulator and study cost-minimizing  ...  DragonFly, SlimFly, and BundleFly topologies.  ...  To simulate network congestion, we inject messages with varying delays by simulating a Poisson process.  ... 
arXiv:2104.11725v2 fatcat:cdavfzaepjaefpbqo2hvlzjbo4

SDNOFS: Software Defined Networking with Openflow Switches & BCN-ECN with ALTQ for Congestion Avoidance

2020 International journal of recent technology and engineering  
Since congestion avoidance is critical for the efficient use of different applications, an efficient method for congestion management in software-defined networks based on Open Flow protocol has been presented  ...  in SDN environments which allows the SDN controller to interact directly with the forwarding plane of network devices such as switches and routers, both physical and virtual (hypervisorbased), so that  ...  While such a device can be remotely manipulated, it is easy to select real machines and networks through simulation.  ... 
doi:10.35940/ijrte.d8873.018520 fatcat:ibe77yniafg7ze4bu656c6qwfy

Application-aware Congestion Mitigation for High-Performance Computing Systems [article]

Archit Patke, Saurabh Jha, Haoran Qiu, Jim Brandt, Ann Gentile, Joe Greenseid, Zbigniew Kalbarczyk, Ravishankar Iyer
2021 arXiv   pre-print
However, the impact of congestion on application runtime differs from application to application depending on their network characteristics (such as bandwidth and latency requirements).  ...  High-performance computing (HPC) systems frequently experience congestion leading to significant application performance variation.  ...  RELATED WORK Modeling impact of congestion on HPC applications. Several researchers have studied the effects of network congestion on the performance of HPC applications.  ... 
arXiv:2012.07755v2 fatcat:braticwr3nhsnmre7hd45rvtvq

FatTreeSim

Ning Liu, Adnan Haider, Xian-He Sun, Dong Jin
2015 Proceedings of the 3rd ACM Conference on SIGSIM-Principles of Advanced Discrete Simulation - SIGSIM-PADS '15  
in data centers and HPC systems, as well as evaluating the applications running on top of the network.  ...  FatTreeSim is designed to model and simulate large-scale fat-tree networks up to millions of nodes with protocol-level fidelity.  ...  This also research used resources of the Argonne Leadership Computing Facility at Argonne National Laboratory, which is supported by the Office of Science of the U.S.  ... 
doi:10.1145/2769458.2769474 dblp:conf/pads/LiuHSJ15 fatcat:avxa3xaje5d4pbi73ixa6wzk4i

Topology-Aware Performance Optimization and Modeling of Adaptive Mesh Refinement Codes for Exascale

Cy P Chan, John D Bachan, Joseph P Kenny, Jeremiah J Wilke, Vincent E Beckner, Ann S Almgren, John B Bell
2016 2016 First International Workshop on Communication Optimizations in HPC (COMHPC)  
Furthermore, we show that network latency in the multigrid bottom solve is the main contributing factor preventing good scaling on exascale-class machines.  ...  and the SSTmacro network simulator.  ...  We analyze two classes of network topologies: the 3D torus and the dragonfly, both modeled by extending the configuration of current interconnects (modifying the topology The on-node performance of the  ... 
doi:10.1109/comhpc.2016.008 dblp:conf/sc/ChanBKWBAB16 fatcat:g3moavjnpfcmfplyazisu2etaa

Topological Characterization of Hamming and Dragonfly Networks and Its Implications on Routing

Cristóbal Camarero, Enrique Vallejo, Ramón Beivide
2014 ACM Transactions on Architecture and Code Optimization (TACO)  
Current HPC and datacenter networks rely on large-radix routers.  ...  Topological characterization of Hamming and dragonfly networks and its implications on routing.  ...  This section explores the properties of 3-level dragonflies.  ... 
doi:10.1145/2677038 fatcat:iudi3oiyj5b5rpz4wvevjhzk3u

Measuring Congestion in High-Performance Datacenter Interconnects

Saurabh Jha, Archit Patke, Jim M. Brandt, Ann C. Gentile, Benjamin Lim, Mike Showerman, Greg Bauer, Larry Kaplan, Zbigniew Kalbarczyk, William Kramer, Ravi K. Iyer
2020 Symposium on Networked Systems Design and Implementation  
on credit-based interconnect networks.  ...  While it is widely acknowledged that network congestion in High Performance Computing (HPC) systems can significantly degrade application performance, there has been little to no quantification of congestion  ...  This research is part of the Blue Waters sustained-petascale computing project, which is supported by the National Science Foundation (awards OCI-0725070 and ACI-1238993)  ... 
dblp:conf/nsdi/JhaPBGLS0KK0I20 fatcat:y5ihsuckxrfd7k5g7g7gxuvtte

Performance optimization of load imbalanced workloads in large scale Dragonfly systems

Bogdan Prisacari, German Rodriguez, Cyriel Minkenberg, Marina Garcia, Enrique Vallejo, Ramon Beivide
2015 2015 IEEE 16th International Conference on High Performance Switching and Routing (HPSR)  
In exploring this vulnerability, we manage to provide a theoretical justification, based on inherent properties of the Dragonfly topology, of why performance degrades.  ...  Dragonfly topologies are one of the most promising interconnect designs for enabling large, potentially exascale compute systems, particularly those envisioned to accommodate workloads that are sensitive  ...  ACKNOWLEDGMENTS This work has been supported by the Spanish Science and Technology Commission (CICYT) under contract TIN2013-46957-C2-2-P, the European HiPEAC Network of Excellence and the JSA no. 2013  ... 
doi:10.1109/hpsr.2015.7483107 dblp:conf/hpsr/PrisacariRMG0B15 fatcat:e7nosvb5qzbxfhk4bqwsass5ly

The Next Generation Of Exascale-Class Systems: The Exanest Project

R. Ammendolay, A. Biagioni, P. Cretaro, O. Frezza, F. Lo Cicero, A. Lonardo, M. Martinelli, P. S. Paolucci, E. Pastorelli, F. Simula, P. Vicini, G. Taffoni (+6 others)
2017 Zenodo  
Furthermore, the consortium is to provide real HPC applications to validate the system. Herein we provide a status report of the project initial developments.  ...  The ExaNeSt project started on December 2015 and is funded by EU H2020 research framework (call H2020-FETHPC-2014, n. 671553) to study the adoption of low-cost, Linux-based power-efficient 64-bit ARM processors  ...  Moreover, a set of configuration/status registers is accessible on the same AXI bus through the "Target Controller" IP, which allows configuration of the router (e.g. setting coordinates and lattice size  ... 
doi:10.5281/zenodo.823594 fatcat:td67qi6nibfl3ozhafhjk2ya7u

ETP4HPC's Strategic Research Agenda for High-Performance Computing in Europe 4 [article]

Michael Malms, Marcin Ostasz, Maike Gilliot, Pascale Bernier-Bruna, Laurent Cargemel, Estela Suarez, Herbert Cornelius, Marc Duranton, Benny Koren, Pascale Rosse-Laurent, María S. Pérez-Hernández, Manolis Marazakis (+11 others)
2020 Zenodo  
This new concept well reflects the main trend of this SRA – it is not only about developing HPC technology in order to build competitive European HPC systems but also about making our HPC solutions work  ...  The main objective of this SRA is to identify the European technology research priorities in the area of HPC and High-Performance Data Analytics (HPDA), which should be used by EuroHPC to build its 2021  ...  class and their control is more efficient.  ... 
doi:10.5281/zenodo.4605343 fatcat:lcsgbea5dzgdfmj5dkw6pr7vni

Rack-scale Computing (Dagstuhl Seminar 15421)

Babak Falsafi, Tim Harris, Dushyanth Narayanan, David A. Patterson, Marc Herbstritt
2016 Dagstuhl Reports  
In addition to stimulating interaction during the seminar, the event led to a follow-on Workshop on Rack-Scale Computing to be organized during 2016.  ...  This report documents the program and the outcomes of Dagstuhl Seminar 15421 "Rack-scale Computing".  ...  It would be interesting to investigate what happens with libfabric (a commonly used HPC library for communication) in face of partial failures. Topologies HPC networks focus on cheaper topologies.  ... 
doi:10.4230/dagrep.5.10.35 dblp:journals/dagstuhl-reports/FalsafiHNP15 fatcat:n2i2d5gh5zc4bbvjz7q4iborki

A Novel Query Caching Scheme for Dynamic InfiniBand Subnets

Evangelos Tasoulas, Ernst Gunnar Gran, Bjorn Dag Johnsen, Tor Skeie
2015 2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing  
The advent of the Internet of Things, sensor and social networks, to mention just a few examples, all contribute towards the solid establishment of the Big Data era.  ...  High Performance Computing (HPC) becomes necessary for the efficient processing of the massive amounts of data our society generates, and cloud computing is a critical component to deliver this processing  ...  Oblivious Routing Congestion Simulator The Oblivious Routing Congestion Simulator (ORCS) [51] is capable of simulating a variety of communication patterns on statically routed networks with the intent  ... 
doi:10.1109/ccgrid.2015.10 dblp:conf/ccgrid/TasoulasGJS15 fatcat:hsxswl63gfbtbdipuhvrhx6y7u
« Previous Showing results 1 — 15 out of 43 results