Blog

Posts Tagged ‘Phoenix’

Phoenix Phase D to Analyse LHC Data Inaugurated

Monday, March 14th, 2011

Today, CSCS announced the inauguration of phase D of the «Phoenix» cluster. Phase D increases the computing capabilities of Phoenix by 20% with 240 more cores, and storage increased by 30% with 230 TB, reaching a total of 1392 cores and 1 PB of disk.

Phoenix is using the data produced at the LHC particle collider at CERN for the Swiss Institute for Particle Physics (CHIPP), analyzing real collisions and producing simulations for the experiments: ATLAS, CMS and LHCb.

In 2010 Phoenix ran more than 2.8 million jobs for more than 5.3 million hours (2009 was 1.7 million jobs), and is expected to increase every year. Christoph  Grab from CHIPP comments as follow the results of the previous Phase C «Various physics analyses have directly profited from the reliable operation of the PHOENIX cluster. Among the most interesting results were measurements of top-quark, W and Y – production as well as detailed studies of the dynamic properties in b-hadron production».

The extension of Phoenix with compute nodes and storage:

 

 

Record CPU Time in September on Phoenix (LCG2 Grid Cluster)

Saturday, October 2nd, 2010

After upgrading to Phase C and making several improvements during the last months, September 2010 surpassed all Phoenix GRID Cluster records running around 200.000 jobs and computing more than 850.000 CPU hours (normalized to 1 CPU = 1000 SpecInt2000) – doubling the average numbers since 2009.

This is really great news for the CHIPP community and CSCS sysadmins, and encourages all to continue with the good work.

This is just one of many breakthroughs to come!

We remember the Phoenix is a Grid Cluster run by CSCS for the Swiss Institute of Particle Physics (CHIPP). Phoenix has been upgraded early this year (read more).

Have a look to this previous posting, if you want to know what the particle physics researcher are doing with Phoenix.

Phoenix PhaseC Upgrade Passed Acceptance Test

Wednesday, March 31st, 2010

CSCS is happy to announce that  Phoenix PhaseC Upgrade passed today the acceptance test. Phoenix is a HPC system run by CSCS used by the researchers of the Swiss Institute of Particle Physics CHIPP to analyze the data from the LHC experiment at CERN.

The upgrade has been delivered and installed by Sun and will double the computing and storage capacity provided by the previous system (also from Sun). The main characteristics of the new cluster are:

  • 96 Sun X6275 compute nodes with in total 768 cores running at 2.56 GHz (Intel E5540)
  • 7.6 TFlops peak performance
  • Lustre 115 TB usable space with a 7.6 GB/s write speed used as scratch
  • 10 Thors X4540 with480 TB total space (plus 672 TB Thumper recycled from PhaseB) used to store experiment data
  • Infiniband QDR used as interconnect

In the next days the system administrators at CSCS will migrate the middleware from phase B to C. PhaseC should be fully operational on Montay, April 19th. You can follow the next steps of the upgrade on twitter: http://www.twitter.com/phoenix_cscs

The Acceptance Protocol being signed by Dominik Ulmer (General Manager, CSCS) and Klaus Landl (Account Manager, Sun)

Part of the project team posing in front of the PhaseC system.

Today we also had the handover of the system responsibility from Fotis (right on the next picture) to Pablo. We thank Fotis for all the excellent job made so far. He has helped make Phoenix into a great cluster during these last 18 months, together with Jason and Riccardo, making it challenging for others to achieve the same quality. Pablo has an extensive GRID background and has been working with us for two months, during which time he has learned the daily procedures at CSCS-LCG2 grid site and is ready to succeed Fotis as system lead.

Installation of the Phoenix Upgrade for CHIPP

Saturday, February 6th, 2010

In January 2010 started at CSCS the upgrade of the cluster “Phoenix” that is being used as tier 2 by the Swiss Institute of Particle Physics (CHIPP) in the LHC experiment at CERN.  The Swiss commitment in 3 of the 4 large LHC experiments (ATLAS, CMS and LHCb) mandates establishing their own Grid computing infrastructure for performing LHC physics data analysis in Switzerland.

The previous system based on SunBlade 8000 for the Compute Notes, SF_X4200 M2 machines with 2.8 GHz AMD processors for the Service Nodes and SF_X4500 for the Storage Nodes with our ZFS-Solaris Technology has been running now for 3 years and an upgrade and expansion was necessary.

The new HPC system is based on a Gigabit Ethernet network and an additional Infiniband network based on QDR technology with the Sun Datacenter Infiniband Switch 648 and uses Lustre as parallel file system. The worker nodes are base on Sun X6275 blade server, based on the new Intel X5500 processor generation (Nehalem, 2.53 GHz, Quad-Core, 8MBCache).

The hardware has been delivered beginning January, 2010 and will be functional in March, 2010. The old compute nodes will be decomissioned by mid year.

In the next picture you can follow the assembling of the system in the first month.

The truck delivering the hardware beginning of January at a sunny day in Manno.

The boxes with the delivered hardware in the CSCS computing room.

Unloading of the racks that can up to 1’080 kg heavy.

Christoph Grab (CHIPP)  inspects the location of the upgrade of Phoenix (which is next to the existing system).

The first QDR InfiniBand cables (in blue) are connected to the IB fabric. Because they are very fragile, the cables are suspended to the deck and not running in the raised floor.

The sysadmins of Phoenix Jason and Fotis are getting ready to take over the new system.