We are developing the MDGRAPE-4, a special-purpose computer system for molecular dynamics (MD) simulations. MDGRAPE-4 is designed to achieve strong scalability for protein MD simulations through the integration of general-purpose cores, dedicated pipelines, memory banks and network interfaces (NIFs) to create a system on chip (SoC). Each SoC has 64 dedicated pipelines that are used for non-bonded force calculations and run at 0.8 GHz. Additionally, it has 65 Tensilica Xtensa LX cores with single-precision floating-point units that are used for other calculations and run at 0.6 GHz. At peak performance levels, each SoC can evaluate 51.2 G interactions per second. It also has 1.8 MB of embedded shared memory banks and six network units with a peak bandwidth of 7.2 GB s-1 for the three-dimensional torus network. The system consists of 512 (8×8×8) SoCs in total, which are mounted on 64 node modules with eight SoCs. The optical transmitters/receivers are used for internode communication. The expected maximum power consumption is 50 kW. While MDGRAPE-4 software has still been improved, we plan to run MD simulations on MDGRAPE-4 in 2014. The MDGRAPE-4 system will enable long-time molecular dynamics simulations of small systems. It is also useful for multiscale molecular simulations where the particle simulation parts often become bottlenecks.
Description
MDGRAPE-4:special computer for MD | Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences
%0 Journal Article
%1 Ohmura2014MDGRAPE
%A Ohmura, Itta
%A Morimoto, Gentaro
%A Ohno, Yousuke
%A Hasegawa, Aki
%A Taiji, Makoto
%D 2014
%I The Royal Society
%J Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences
%K ASIC computers hardware molecular-dynamics special-built-hardware
%N 2021
%R 10.1098/rsta.2013.0387
%T MDGRAPE-4: a special-purpose computer system for molecular dynamics simulations
%U http://rsta.royalsocietypublishing.org/content/372/2021/20130387
%V 372
%X We are developing the MDGRAPE-4, a special-purpose computer system for molecular dynamics (MD) simulations. MDGRAPE-4 is designed to achieve strong scalability for protein MD simulations through the integration of general-purpose cores, dedicated pipelines, memory banks and network interfaces (NIFs) to create a system on chip (SoC). Each SoC has 64 dedicated pipelines that are used for non-bonded force calculations and run at 0.8 GHz. Additionally, it has 65 Tensilica Xtensa LX cores with single-precision floating-point units that are used for other calculations and run at 0.6 GHz. At peak performance levels, each SoC can evaluate 51.2 G interactions per second. It also has 1.8 MB of embedded shared memory banks and six network units with a peak bandwidth of 7.2 GB s-1 for the three-dimensional torus network. The system consists of 512 (8×8×8) SoCs in total, which are mounted on 64 node modules with eight SoCs. The optical transmitters/receivers are used for internode communication. The expected maximum power consumption is 50 kW. While MDGRAPE-4 software has still been improved, we plan to run MD simulations on MDGRAPE-4 in 2014. The MDGRAPE-4 system will enable long-time molecular dynamics simulations of small systems. It is also useful for multiscale molecular simulations where the particle simulation parts often become bottlenecks.
@article{Ohmura2014MDGRAPE,
abstract = {We are developing the MDGRAPE-4, a special-purpose computer system for molecular dynamics (MD) simulations. MDGRAPE-4 is designed to achieve strong scalability for protein MD simulations through the integration of general-purpose cores, dedicated pipelines, memory banks and network interfaces (NIFs) to create a system on chip (SoC). Each SoC has 64 dedicated pipelines that are used for non-bonded force calculations and run at 0.8 GHz. Additionally, it has 65 Tensilica Xtensa LX cores with single-precision floating-point units that are used for other calculations and run at 0.6 GHz. At peak performance levels, each SoC can evaluate 51.2 G interactions per second. It also has 1.8 MB of embedded shared memory banks and six network units with a peak bandwidth of 7.2 GB s-1 for the three-dimensional torus network. The system consists of 512 (8{\texttimes}8{\texttimes}8) SoCs in total, which are mounted on 64 node modules with eight SoCs. The optical transmitters/receivers are used for internode communication. The expected maximum power consumption is 50 kW. While MDGRAPE-4 software has still been improved, we plan to run MD simulations on MDGRAPE-4 in 2014. The MDGRAPE-4 system will enable long-time molecular dynamics simulations of small systems. It is also useful for multiscale molecular simulations where the particle simulation parts often become bottlenecks.},
added-at = {2018-04-09T16:27:29.000+0200},
author = {Ohmura, Itta and Morimoto, Gentaro and Ohno, Yousuke and Hasegawa, Aki and Taiji, Makoto},
biburl = {https://www.bibsonomy.org/bibtex/2e0d363e13fe3aab299d85db713aa7835/salotz},
description = {MDGRAPE-4:special computer for MD | Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences},
doi = {10.1098/rsta.2013.0387},
eprint = {http://rsta.royalsocietypublishing.org/content/372/2021/20130387.full.pdf},
interhash = {26b703a6bb116de7811c0e142c83cc81},
intrahash = {e0d363e13fe3aab299d85db713aa7835},
issn = {1364-503X},
journal = {Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences},
keywords = {ASIC computers hardware molecular-dynamics special-built-hardware},
number = 2021,
publisher = {The Royal Society},
timestamp = {2018-04-09T16:27:29.000+0200},
title = {MDGRAPE-4: a special-purpose computer system for molecular dynamics simulations},
url = {http://rsta.royalsocietypublishing.org/content/372/2021/20130387},
volume = 372,
year = 2014
}