Archive for January, 2010

Decommission of «Palu» at CSCS

Wednesday, January 27th, 2010

On January 27th, 2010 the Cray XT3 has reached the end of its life here at CSCS, for the last time the system has been shut down and disassembled.

This computer has been named after Piz Palü in the Bernina range in Graubunden, Eastern Switzerland with an elevation 3901 m and his last configuration has been the follow:

  • 6 Cabinets containing 14 service processing elements (PEs), subdivided into 7 service blades, and 548 dual core nodes giving 1096 compute PEs subdivided into 137 compute blade
  • 3 GB of Ram (Compute nodes)

Palu started its life as production system on January 2006. This computer was the very first Cray XT machine to set foot on Europe. It was the first supercomputer based on the XT architecture thatrun Catamount operating system using Infiniband as high-speed interconnect. Palu was mainly used for massively parallel jobs.

Of the 6 cabinets of Palu, two will be disassembled by Cray to be used as spare parts, one will stay at CSCS and be the first exhibit of an own museum. The last three cabinets will be shipped next week to a compter museum near to Solothurn. Stay tuned on this blog to get additional information on the museum next week…

In the next photos you may see the technician of Cray starting the disassembling of Palu (this will need almost three working days).

Disassembling of the interconnect.

The interconnect cables and removing the power supply cables.

Another view of the interconnect being disassembled:

Palu_Interconnect

Disassembling part of the compute blades to be used as spare parts.

Move of Dôle at CSCS to a New Location

Tuesday, January 26th, 2010

Today the HPC system called “Dôle” has been moved to a new location inside CSCS. Dôle is being used by MeteoSwiss as failover for Buin, the main production system for weather forecast.

Previously Dôle and Buin were placed next to each one. The replacement to two separate locations in the CSCS computer room will ensure a maximum of availability in case of any technical issue.

Case Study of University of Zurich and Intel on High-Performance Computing

Monday, January 25th, 2010

University of Zurich and Intel commonly announced a case study for the use of Intel® Xeon® processor 5500 and 5400 series for Schrödiger, the new Sun HPC system of the university.

UniZh_Intel

The University has been at the forefront of scientific research for many years and relies heavily on its HPC cluster to underpin complex calculations and simulations. Its environment was beginning to age, resulting in slow response times and even the inability to carry out certain simulations.

Dr. Alexander Godknecht, head of IT-infrastructure, bioinformatics and HPCN, IT Services at the University of Zürich, explains: “Many of our compute-heavy departments were having trouble getting what they needed out of the old platform. The astrophysics team, for example, needs large amounts of memory to carry out its calculations while the physical chemists require fast networks with low latency and multiple cores in order to get the compute performance to support their computations. Meanwhile, the biochemistry researchers were hardly able to compute their thousands of simulations as the time taken to do them was just too long.”

University of Zurich selected for the new HPC system called Schrödinger a solution provided by Sun and Intel. The environment deployed is underpinned by six 48-blade racks of Sun Blade X6275 server modules, powered by a total of 4,608 Intel Xeon processor 5500 series cores. Running a SUSE Linux Enterprise* operating system, it supports all the applications used by the various departments to ensure the cluster is kept free to run the parallel applications for which the Intel Xeon processor 5500 series is optimised.

“We’re confident that the new cluster provided by Sun and Intel will last us for a good few years and enable us to push ahead with new scientific breakthroughs that Schrödinger himself would be proud of,” concludes Dr. Godknecht. The industry has already recognised the University’s new cluster by ranking it 96th in the Top 500 Supercomputers worldwide.

While the cluster is currently used exclusively by scientists based at the University of Zürich, it forms a part of the general HPC strategy in Switzerland led by the Swiss National Supercomputing Centre (CSCS). Like other countries, Switzerland has a strategy for a national HPC infrastructure. A grid or a series of smaller clusters form the base of the pyramid, followed by big clusters like Schrödinger and at the top of the national pyramid will be the planned Petabytelevel Supercomputer at CSCS. By providing a platform where scientists can write and test codes for thousands of processing cores, the University of Zürich will be part of the Swiss national plan for High Performance Computing and Networking.

Press Release: Case Study Univerity of Zurich and Intel »

HPC @ University of Bern: ubelix – Uni BErn LInuX cluster

Friday, January 22nd, 2010

See also the original slides of Andres Aeschlimann as pdf »

Purpose
This Grid HPC infrastructure is primarily designed to support the researchers at the Campus. They should use their time doing research and not be bothered by deploying a Grid HPC infrastructure.

Picture of ubelix

Some facts

  • first Linux Cluster was installed in 2001 (1 master and 32 single core nodes)
  • continuously expanded to ~1000 cores in >200 nodes today
  • Dual- and quadcore worker nodes
  • Mostly Opterons, increasing # of Intels (Nehalem)
  • several suppliers (mostly SUN, but currently also IBM and some Dell)
  • < 100kW
  • Gentoo Linux www.gentoo.org
  • Kernel 2.6.22/2.6.27
  • 2TB memory, 50TB disk
  • Lustre filesystem: 1.8.1
  • Sun Grid Engine 6.2
  • Gb Switch
  • Currently no Infiniband Switch

Internal (private) network

  • TCP/IP
  • Stackable Switches (~40Gbs)
  • „normal“ Gigabit Ethernet on the worker nodes
  • 10GE Ethernet for high throughput servers

Internal Network of ubelix

Lustre@ubelix

ubelix Lustre

Application portfolio (local users)

  • HE Physics
  • Astronomy
  • Computational and Molecular Population Genetics Lab
  • Space Research Physics
  • Computer Vision and Artificial Intelligence
  • Chemistry and Biochemistry

Applications from remote (SMSCG)

  • ATLAS: high energy physics application developed for the LHC experiment at CERN
  • RSA768: cryptographic application
  • NAMD and GROMACS: biochemistry applications
  • GAMESS: biochemistry application (work in progress)

Remote access to cluster

Other clusters @ UniBE

  • The LHEP UNIBE Atlas T3 2009 – A ROCKS Cluster with ~200 cores (Sun Fire X2200 IU dual quad cores) and ~50 TB on CentOS. Located in same room as ID UNIBE clusgter. Mainly serves local and remote ATLAS scientists. Backfilled with remote users and applications. Speciality: Access only via ARC clients, i.e. remote and local users habe the same interface. http://ce.lhep.unibe.ch
  • Theoretical Physics (~200 cores, with interconnect)
  • Climate Physics (~100 cores)
  • Space Physics (~100 cores)
  • Chemistry (~100 cores, with interconnect)
  • Computational and Molecular Population Genetics (~60 cores)

Handbook of Research on Computational Science and Engineering: Call for Proposals

Thursday, January 21st, 2010

We are pleased to invite you to submit your proposals for the contribution of chapters to the “Handbook of Research on Computational Science and Engineering: Theory and Practice”.

CSE is an emerging, rapidly developing, and potentially very significant force in changing scientific practice by offering a ‘third way’ of carrying out research in addition to, or indeed, instead of, theory and experiment.

lGI_Global_ogo

This handbook will provide a basic reference text for the fundamental elements making up CSE and showing their interdependence in a way that (a) reviews state of the art and  current achievements; (b) explores imminent developments that will advance the state of play; and (c) presents these in a form accessible to as wide an audience of interested parties as possible.

Details are provided on the book site http://www.cse-book.com/index.html and on the following PDF document: CSE-handbook-call-extended.

Mario Valle (CSCS) on behalf of the Editorial Board