Difference between revisions of "QUonG initiative"

From APEWiki
Jump to: navigation, search
Line 1: Line 1:
==The '''QUonG''' Project==
+
==The '''QUonG''' Initiative==
  
 
[[File:Elemento_QUonG.png|border|left|300px]]  
 
[[File:Elemento_QUonG.png|border|left|300px]]  
Line 18: Line 18:
 
The network is built upon the [[APEnet+_project|APEnet+]] project.
 
The network is built upon the [[APEnet+_project|APEnet+]] project.
  
=== Upcoming system ===
+
=== Prototype system ===
 +
The first prototype of the '''QUonG''' parallel system is expected to
 +
be delivered before the end of 2011. The chosen ratio for this first
 +
deliverable is one (multi-core) host for two GPU's.
  
A first '''QUonG''' system prototype is expected to be delivered at the
+
As a consequence, the '''QUonG''' elementary mechanical assembly is a
end of the year 2011 while from the beginning of 2012 we will be ready
+
3U "sandwich" made of two Intel-based servers plus a NVIDIA S2050
to start mass production of '''QUonG'''.
+
multiple GPU system equipped with the APEnet+ network board.
 +
 
 +
The assembly collects two '''QUonG''' elementary computing units, each
 +
made of one server hosting the interface board to control 2 (out of 4)
 +
GPU's inside the S2050 and one APEnet+ board. In this way, a
 +
'''QUonG''' elementary computing unit is topologically equivalent to
 +
two vertexes of the APEnet+ 3-dim mesh.
 +
 
 +
=== Envisioned upcoming system ===
  
 
Building from the configuration of the first '''QUonG''' prototype
 
Building from the configuration of the first '''QUonG''' prototype
equipped with a current NVIDIA Tesla GPU with Fermi architecture, we
+
equipped with current NVIDIA Tesla GPU's with Fermi architecture, we
 
envision a final shape for a deployed '''QUonG''' system as an assembly
 
envision a final shape for a deployed '''QUonG''' system as an assembly
 
of fully populated standard 42U height '''QUonG''' racks, each one
 
of fully populated standard 42U height '''QUonG''' racks, each one

Revision as of 15:24, 21 June 2011

The QUonG Initiative

Elemento QUonG.png
QUonG rack.png

QUonG is an Logo-infn.pngINFN (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 latest generation GPUs; its communication mesh - being tailored to the characteristics of LQCD algorithms - is a point-to-point, high performance, low latency network where the computing nodes are topologically arranged as vertexes of a 3 dimensional torus.

The network is built upon the APEnet+ project.

Prototype system

The first prototype of the QUonG parallel system is expected to be delivered before the end of 2011. The chosen ratio for this first deliverable is one (multi-core) host for two GPU's.

As a consequence, the QUonG elementary mechanical assembly is a 3U "sandwich" made of two Intel-based servers plus a NVIDIA S2050 multiple GPU system equipped with the APEnet+ network board.

The assembly collects two QUonG elementary computing units, each made of one server hosting the interface board to control 2 (out of 4) GPU's inside the S2050 and one APEnet+ board. In this way, a QUonG elementary computing unit is topologically equivalent to two vertexes of the APEnet+ 3-dim mesh.

Envisioned upcoming system

Building from the configuration of the first QUonG prototype equipped with current NVIDIA Tesla GPU's with Fermi architecture, we envision a final shape for a deployed QUonG system as an assembly of fully populated standard 42U height QUonG racks, each one capable of 60 TFlops/rack in single precision (30 TFlops/rack in double precision) of peak performance, at a cost of 5kEuro/TFlops and for an estimated power consumption of 25 kW/rack.

A typical QUonG PFlops scale installation, composed of 16 QUonG racks interconnected via APEnet+ 3-dim torus network, will be characterized by an estimated power consumption of the order of 400 kW and a cost ratio of the order of 5 kEuro/TFlops.