Blog

Archive for May, 2010

ETH Zurich and IBM Rüschlikon Announce Aquasar

Saturday, May 15th, 2010

ETH Zurich and IBM Rüschlikon announced on May 6th that Aquasar is operational. Aquasar is an HPC system developed together by the two institutions using water to directly cool the integrated circuits. The water with a temperature of about 60 oC is used to heat the building of ETH Zurich. The goal of this research project is to reduce the energy footprint of computing systems: It is assumed that computers use about 5 to 10% of the electricity worldwide.

Prof. Ralph Eichler, President of ETH Zurich and Dr. John Kelly, Senior Vice President IBM Research, present Aquasar. (Foto: Michael Lowry, IBM Research – Zurich)

Aquasar has a computing power of 6 Teraflops and consumes about 20 kilowatt of electricity. Water cooling on the chip may be the big next step to build larger supercomputers and to go to Exaflops.

Read more about Aquasar on ETH Life »
or read the press release of IBM Research »

Agenda hpc-ch Forum on “Operating Systems for HPC and Installation Methods”

Thursday, May 13th, 2010

We are  happy to announce the final agenda of the second Forum of the hpc-ch community on “Operating Systems for HPC and Installation Methods” (EPF Lausanne, May 20).

In total we have 31 registrations, 6 presentations and 2 site visits at the infrastructure of EPF Lausanne. We would like Vittoria Rezzonico for the big help in organizing this day.

See the detailed agenda »

Attention – Change of restaurant: The hpc-ch meeting will already start on Wednesday, May 19th with a common dinner starting at 19:00 at the Pizzeria le Milan, Boulevard de Grancy 54, 1006 Lausanne.

10:00 – 10:30 Registration and welcome co ffee

10:30 – 10:45 Greeting and introduction

  • Welcome note by Jean-Claude Berney and Didier Rey (EPF Lausanne)
  • Introduction by Vittoria Rezzonico (EPF Lausanne)

10:50 – 11:30 Keynote presentations

  • AutoYaST and Installation of central services clusters; Jacques Menu
    (EPF Lausanne)
  • Installation of a cluster of PS3s; Maxime Augier (EPF Lausanne)

11:30 – 11:50 Data and computing centers at EPFL, a brief introduction

11:50 – 12:10 Site visit of data and computing centers, Central Services + FSB

12:10 – 13:00 Bu ffet lunch

13:00 – 13:40 Site visit of data and computing centers, Central Services + FSB

13:45 – 14:30 Site visit of the I&C computing infrastructures (PlayStations!)

14:30 – 15:15 Presentations by hpc-ch members (part one)

  • Configuration management with puppet and coupling it to kickstart at PSI; Valeri Markushin (PSI)
  • An overview of operating systems, installation procedures and monitoring at CSCS; Peter Öttl and Fotis Georgatos (CSCS)

15:15 – 15:45 Tea and Coff ee break

15:45 – 16:30 Presentations by hpc-ch members (part two)

  • Experiences with the Rocks Distribution for Cluster Deployment at USI; Dorian Krause (USI)
  • Managing the Bernese cluster using the tools from gentoo.org; Stefan Ott (University of Bern)

16:30 – 17:00 Final discussion and farewell

New Visualization Cluster Eiger Delivered at CSCS

Saturday, May 8th, 2010

The Visualization/Research & Development Cluster EIGER is a new CSCS facility which extends the current resource portfolio. During the Q2/Q3 2010 it will be fully integrated into the CSCS Supercomputing ecosystem, and will be opened to the swiss scientific user community for hybrid multicore/ multi-GPU computing, visualization, data analysis, and general purpose pre/post processing activities.

The Eiger mountain in the Swiss Alps …

… and the Eiger cluster in the CSCS machine room being assembled (the interconnect cables are still missing).

The EIGER cluster is a tightly coupled computing cluster system, running Novell SUSE Linux Enterprise Server 11 Operating System release and includes 19 nodes based on the dual-socket six-cores AMD Opteron 2427 processor architecture running at 2.2 GHz, offering 24 GB of main system memory per node, for a total of 228 cpu cores and 552 GB aggregate memory. 4 out of 19 cluster nodes offers a larger main system memory capacity up to 48 GB.

Altair PBS Professional V 10.3 is the main batch queuing system installed and supported on the cluster in order to let end-users access in a shared or reserved mode any available visualization/computing resource.

Several class of nodes have been defined inside the cluster, covering special functionalities :

  • Class 0: Administration Node (1x)
  • Class 1: Login Node (1x)
  • Class 2: Visualization Nodes (7x)
  • Class 3: Fat Visualization Nodes (4x)
  • Class 4: Advanced Development Nodes (4x)
  • Class 5: Storage Nodes (2x)

Depending on the node class membership, cluster nodes are equipped with one of the two kind of GPUs family products :

  • NVIDIA GeForce GTX 285 2 GB => Class 2/3 – soon to be extended with NVIDIA GeForce GTX 480 1.5 GB
  • NVIDIA TESLA S1070 GPUs 4 GB => Class 4 – soon to be extended with 2 upcoming new NVIDIA TESLA/FERMI S2070 6 GB

As an high speed network interconnect, the cluster EIGER rely on a dedicated Infiniband QDR fabric infrastructure, supporting both parallel-MPI traffic and the internal parallel scratch file system I/O data traffic. In addition, a commodity 10 GbE LAN ensures interactive login access, home, project and application file sharing among the cluster nodes, and a standard 1 Gbe administration network is also reserved for cluster management purposes.

CSCS Training Course Reminder – May/June 2010

Wednesday, May 5th, 2010

The Swiss National Supercomputing Centre (CSCS) will be hosting the following training courses during the months of May and June 2010. We still have some places available in each course so we encourage you to register quickly to benefit from these opportunities.

A “Hands-On” Introduction to PETSc (10–11/05/2010)
This course provides students with a comprehensive introduction to the highly-scalable PETSc mathematical library which is ideally suited to the rapid development of scientific applications modeled by partial differential equations. The course is further complemented by an overview of other public-domain parallel mathematical libraries routinely used within the HPC community. This course will be presented by Jed Brown (ETHZ and official PETSc Developer) and Will Sawyer (CSCS).

Please register here for this course »

The course itself is free of charge, however, attendees must cover their own travel and accommodation expenses. The deadline for this event is rapidly approaching, so please register quickly.

Rosa Introduction Course (14–15/06/2010)
This introduction course will provide both existing and new users with an introduction to the Rosa system including detailed information regarding the Cray XT5 hardware, software architectures and programming environments. Furthermore, users will be gain practical experience with high-level profiling and optimization techniques on the new system, through a series of hands-on exercises.

Please register here for this course »

The deadline for this event is the *17th of May*. We recommend you to register as soon as possible. The course itself is free of charge, however, attendees must cover their own travel and accommodation expenses. You may want to combine your attendance of this course with the course on “Scalable performance analysis of large-scale parallel applications” that will take place at CSCS on June 16th/17th.

Scalable Performance Analysis of Large-Scale Parallel Applications (16–17/06/2010)
An introduction course to the Scalasca performance analysis toolset which will be run in collaboration with our colleagues fromJülich Supercomputer Centre, who are the main developers of this tool.

Scalasca is an open-source toolset that can be used to analyze the performance behavior of parallel applications and to identify opportunities for optimization. It has been specifically designed for use on large-scale systems including IBM Blue Gene and Cray XT, but is also well-suited for small- and medium-scale HPC platforms.

Scalasca integrates runtime summaries with in-depth studies of concurrent behavior via event tracing. A distinctive feature is the ability to identify wait states that occur, for example, as a result of unevenly distributed workloads. This course will provide the unique opportunity to interact directly with experts from the Jülich Research Centre.

Click here to register for this course »

The deadline for registering for this event is the 17th of May.

We look forward to meeting you at these events.