HLRN System News

Last update: 08 November 2019, 17:26


IDPublished Subject
[3446] Nov 08, 2019Data migration Berlin: update re. final phase
[3444] Oct 31, 2019Course announcements
[3434] Oct 16, 2019Bye bye Konrad
[3422] Oct 02, 2019Apply for Computing Time by October 28th, 2019
[3414] Sep 11, 2019New GPU node in Göttingen
[3400] Aug 05, 2019Data migration from Cray storage to the new HLRN-IV storage started
[3394] Jul 29, 2019External target host handling revised, effective August 5.
[3391] Jul 09, 2019Action required: HLRN-IV installation schedule
[3389] Jul 05, 2019Emmy and Lise - Names for the HLRN-IV compute systems revealed


Data migration Berlin: update re. final phase
[3446] Nov 08, 2019

*** You should take immediate action if you have not started yet ***

Update regarding HLRN in Berlin:

  1. CW47 - CW49:
    Data migration final phase now from Nov 18 until Dec 08, 2019, see also
    https://support.hlrn.de/home/view/NewsCenter/HLRN4News

  2. General announcement:

    Dear HLRN Users,

    since August 5, 2019, you can and have to migrate your data
    in Berlin from the old HLRN-III (Cray) home and work directories
    to the new HLRN-IV filesystems in Berlin yourselves.

    After the end of the data migration period on Dec. 08, 2019,
    the old file systems will no longer be available and all data
    that has not been transferred by you will be lost.

    For more details and examples for efficient data transfer see
    https://www.hlrn.de/home/view/System4/DataMigration .

(wwb)



Course announcements
[3444] Oct 31, 2019
Dear HLRN users,

we would like to announce the following courses taking place in Göttingen:

Node-Level Performance Engineering
Dr.-Ing. Jan Eitzinger, Regionales Rechenzentrum Erlangen (RRZE), FAU Erlangen-Nürnberg
November 19-20, 2019

Parallelization with MPI and OpenMP
Dr. Rolf Rabenseifner (Stuttgart, member of the MPI-2/3/4 forum)
November 25-27, 2019

For more details and registration, please visit the HLRN event calendar

Please also forward this announcement to colleagues who might be interested.

(ck)


Bye bye Konrad
[3434] Oct 16, 2019

The operation of the Berlin Cray system Konrad has now been stopped forever, making place for its successor, the HLRN-IV system Lise, which will be available for general user access in December.

In the last years, the HLRN-III systems Konrad and Gottfried helped the scientific community in North Germany and beyond to solve challenging problems. Since 2014 we provided 2 billion core hours in 2.6 million jobs. With the Cray architecture, we were happy to sustain an availability of more than 98% over the last years.

Thank you for using the HLRN-III resources!

See you on the HLRN-IV systems Lise and Emmy.

The HLRN Team



Apply for Computing Time by October 28th, 2019
[3422] Oct 02, 2019

The Scientific Board of the HLRN is inviting project proposals applying for computing time on the HLRN system.

The next deadline is October 28th, 2019, at 23:59.

Projects that need more computing time than 10,000 NPL resource units per quarter are invited to submit a project proposal ("Großprojektantrag") to the Scientific Board ("Wissenschaftlicher Ausschuss") of HLRN for the next quarterly review. Resources are allocated for one year starting January 1st, 2020, on a quarterly basis after review of the proposal (see the Application HowTo and the Scientific Board portal page). Notifications will be sent out around mid-December 2019.

In 2019 HLRN allocates about 135 million core hours per quarter (equivalent to about 60 million NPL per year) to successful proposals.
Please note: The project abstract has to be created in a specific format. The appropriate HLRN LaTeX template can be downloaded from the HLRN web page "required documents". The use of this template is required.

Please contact your HLRN project consultant before submitting the proposal.

(wwb)



New GPU node in Göttingen
[3414] Sep 11, 2019
Dear users,

we are happy to announce that the first GPU node went into production in Göttingen today. The node (ggpu01) has the following hardware:
2 x Intel Skylake Gold 6148 CPUs (40 cores per node)
192 GB memory
1 x 480 GB solid-state disk
1 x Intel Omni-Path host fabric adapter
4 x NVIDIA Tesla V100 GPUs

This node will be a shared node for data pre- and postprocessing as well as machine learning. The CUDA and cuDNN libraries are installed on the node and a GPU accelerated TensorFlow environment has been added to the anaconda3/2019.03 module. More information on how to use the GPUs can be found in our documentation.

(mb)



Data migration from Cray storage to the new HLRN-IV storage started
[3400] Aug 05, 2019

Dear HLRN Users,

starting immediately you can and have to migrate your data from the old HLRN-III (Cray) home and work directories to the new HLRN-IV filesystems.

After the end of the data migration period on Nov 24, 2019, the old file systems will no longer be available and all data that has not been transferred by you will be lost.

For more details and examples for efficient data transfer see our data migration page.

(tw/wwb/cs)



External target host handling revised, effective August 5.
[3394] Jul 29, 2019

Action required!

Dear users,

from August 5, 2019, on HLRN will open outgoing network connections world wide, however limited to destination ports for ssh, http, and https (22, 80, and 443). At the same time HLRN will remove the list of external target hosts, that a number of users have maintained, as this is not needed anymore.

Action is required for users, that need to connect to external servers in their workflow via non-standard ports. This includes

- users of commercially licensed software (Abaqus, Ansys, CFX, StarCCM+, etc.),

- users connecting to external ports other that the above mentioned.

Those users, and others, that experience connection problems in their workflow on HLRN after August 5, should immediately contact HLRN Support indicating the external host (hostname and ip address) and the port(s), that need to be reached from HLRN.

(wwb)



Action required: HLRN-IV installation schedule
[3391] Jul 09, 2019

Important dates:

August 5th, 2019: Begin of the user initiated data migration phase in Berlin

October 16th, 2019: Cray XC40/30 is powered off forever


Dear HLRN users,

we want inform you about the next major steps in preparing the operation of the phase 2 of the HLRN-IV complexes in Berlin and Göttingen. To our regret, due to the high power demands for the computer systems and cooling it is technically impossible for us to make a smooth transition in Berlin from the old to the new HLRN system, and in Göttingen from the phase 1 to the full system, respectively.

With the start of the test period of the new HLRN-IV configuration in Berlin the compute resource of the old HLRN-III system (Cray) need to be switched off (forever). Operation of the HLRN-IV system in Göttingen will be interrupted to integrate phase 1 and phase 2.

In the following, we provide you with the best technical solution for the transition to the new HLRN-IV system. Of course, the HLRN team will be happy to give you further information and to assist you in this transition process.

You will find more details and regularly updated information about the installation of the phase 2 of the HLRN-IV system at

https://www.hlrn.de/home/view/NewsCenter/HLRN4News

(tst/wwb)



Emmy and Lise - Names for the HLRN-IV compute systems revealed
[3389] Jul 05, 2019

HLRN has revealed the names of the HLRN-IV systems: Lise in Berlin, and Emmy in Göttingen. For details about the namesakes of HLRN-IV please see the corresponding article in our NewsCenter.

(wwb)


For details of short news and older entries see the HLRN System - News Archive


This topic: NewsCenter > WebHome > News
Topic revision: r3 - 2008-08-26 - StefanWollny
 
Back to top of page