Difference between revisions of "QUonG initiative"
From APEWiki
(Created page with ''''QUonG''' is our proposal for a new generation of hybrid GPU-CPU HPC cluster dedicated (but not limited) to LQCD simulations. ==Project Background== Many scientific computatio...') |
|||
| Line 1: | Line 1: | ||
| − | '''QUonG''' | + | ==The '''QUonG''' Project== |
| − | + | '''QUonG''' is an INFN (Istituto Nazionale di Fisica Nucleare) initiative | |
| − | + | targeted to develop a High Performance Computing system dedicated (but not | |
| − | + | limited) to [http://en.wikipedia.org/wiki/Lattice_qcd Lattice QCD] computations. | |
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | + | '''QUonG''' is a massively parallel computing platform built up from | |
| − | + | commodity multi-core processors coupled with last generation GPUs; | |
| − | + | exploits the characteristics of LQCD algorithms, its network mesh is a | |
| − | + | point-to-point, high performance, low latency 3 dimensional toroidal | |
| − | + | network interconnecting the computing nodes. | |
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | + | The network is built upon the [[APEnet+_project|APEnet+]] project. | |
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| + | The final shape of a deployed '''QUonG''' system is an assembly of | ||
| + | standard 42U racks, each one capable of 60 TFlops/rack of peak | ||
| + | performance, at a cost of 5kEuro/TFlops and for an estimated power | ||
| + | consumption of 25 kW/rack. | ||
| − | + | A first '''QUonG''' system prototype is expected to be delivered at the | |
| − | + | end of the year 2011. | |
| − | |||
| − | + | ===Highlights=== | |
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
Revision as of 10:15, 21 June 2011
The QUonG Project
QUonG is an INFN (Istituto Nazionale di Fisica Nucleare) initiative targeted to develop a High Performance Computing system dedicated (but not limited) to Lattice QCD computations.
QUonG is a massively parallel computing platform built up from commodity multi-core processors coupled with last generation GPUs; exploits the characteristics of LQCD algorithms, its network mesh is a point-to-point, high performance, low latency 3 dimensional toroidal network interconnecting the computing nodes.
The network is built upon the APEnet+ project.
The final shape of a deployed QUonG system is an assembly of standard 42U racks, each one capable of 60 TFlops/rack of peak performance, at a cost of 5kEuro/TFlops and for an estimated power consumption of 25 kW/rack.
A first QUonG system prototype is expected to be delivered at the end of the year 2011.
