Dear users,

the HEPMDB interface to IRIDIS HPC cluster is temporarily unavailable. Sorry for the inconvenience.
In case of any questions or problems please contact us at hepmdb@soton.ac.uk
or ask question at launchpad ttps://answers.launchpad.net/hepmdb/+addquestion
or see the answers on other users questions at launchpad https://answers.launchpad.net/hepmdb

Regards
HEPMDB team

Dear all, we are happy to announce that PhenoData API is now available! You can use our API to access endpoints, which allows you to upload data and view papers stored in the database in non-interactive regime. We have language bindings in Shell, and further examples in Python and Javascript (AJAX) The base API endpoint is the URL http://hepmdb.soton.ac.uk/phenodata/api/public

Using API you can upload a lot of data avoiding Interactive upload via Web Interface. The example of the bash script for the upload of the data is available at this link http://hepmdb.soton.ac.uk/phenodata/api/#upload-a-paper
together with the example of the data file.

Dear Friends,

as a result of discussion at the last (Re)Interpretation workshop
http://indico.cern.ch/event/571190/ we have created PhenoData database
and happy to announce its start now.

Here is the link: https://hepmdb.soton.ac.uk/phenodata/. (also can be seen at home page of HEPMDB).

The idea of PhenoData, the part of the HEPMDB project, is to centrally and effectively store data (digitized curves from figures, tables etc) from those HEP papers which did not provide data in arXiv or HEPData, and to avoid duplication of work of HEP researchers on digitizing plots. This database has an easy search interface and paper identification via arXiv, DOI or preprint numbers. PhenoData is not intended to be a replication of any existing archive.

You may view and download existing PhenoData records without registration. You can also upload your own data, to do this you must register at HEPMDB. PhenoData allows you to created new record or update an existing one. Please note that PhenoData is in the development stage. You are welcome to contact us with your questions/suggestions using link at the top of the page or at the launchpad.

PhenoData Developers: Alexander Belyaev, James Blandford, Daniel Locke

Hi all,
There were a few site-breaking bugs fixed, if your account is broken due to blank model names, please reset your account by going Calculate -> Menu -> Reset Account, and it will repopulate the blank models and fix the issues raised.
A big search bug was fixed, and now the website's search function works correctly, hopefully allowing you to explore the Model Database a LOT easier!

If you find any more of these issues, please report them via Launchpad or email us from the "About" page. Please report to us your user ID, too - it makes reproducing these bugs MUCH easier!

The upgrade to the new supercluster has been completed! All user data has been migrated, no data has been lost.

IRIDIS4 has a raw peak performance of ~250 TFLOPS (terra floating-point operations per second) compared with the ~110 TFLOPS of IRIDIS3 - quite a sizeable increase in performance! With this, a redesigned website for HEPMDB has been created, bringing a large amount of needed bug fixes and heavy usability increases.

HEPMDB will be down between the 16th and 18th of July for a planned HPCx migration from IRIDIS3 to IRIDIS4, bringing huge performance gains for our users. This will greatly improve performance for all calculations! Jobs running during the downtime will be removed, so we suggest that you wait for the upgrade to finish to use the new supercluster.

Saving Data from your plots is available now!
Once you produced your plot at "Calculate" session, in addition to the old options to save figure in .PDF, .EPS, .JPG
you have now an option to save Plot Data into file to use it for off-line plots/analysis which gives you a lot of freedom and opportunities. For example, using several data files you can, produce superimposed plots for the sake of comparison etc.
Regards,
HEPMDB team

Dear HEPMDB users, running of Jobs will be not available at HEPMDB in the period 1-4 th of July
due to the maintenance.
Sorry for the inconvenience and thanks for your understanding.
HEPMDB team.

HEPMDB Maintenance today, March 17, 2014, 19:00-24:00 UK time
There will be HEPMDB Maintenance today, March 17, 2014, 19:00-24:00 UK time when the HEPMDB service will be not available. Sorry for inconvenience, HEPMDB Team.

There will be HEPMDB Maintenance tomorrow,
January 24, 2014 during 9:30-10:30 UK time
when the HEPMDB service will be not available.

Sorry for inconvenience,
HEPMDB Team.

CalcHEP batch was updated to prevent instability when performing multidimensional scan over the parameter space.
This instability was very rare, so finally we have pin down the reason [thanks to Darin Baumgartel for intensive usage and test of HEPMDB]
and have fixed the batch system [thanks to Neil Christensen].

If you have experienced the problem when running scan, please, clear the library with matrix elements
which could be corrupted. To do this, go to "Calculate" section and then chose "CalcHEP" package and then chose "Clear Processes" from the Menu. It could take up to several minutes to "Clear Processes" [depending on your activity], so be patient.
Once this operation is finished, you can run your jobs normally.

The running of CalcHEP in the batch regime was improved and now it includes the option to scan over the collider energy.

You can scan over a parameter that is not a model parameter. This would allow you to scan over the collider energy. A lot of thanks to Neil Christensen!

For example, you could add the lines:

Run parameter: rtS
Run begin: 7
Run step size: 1
Run n steps: 2


to your batch file even though rtS is not a parameter of the model. Then the momenta for colliding particles can be made functions of the scan parameters. For example,

p1: 1000*rtS/2
p2: 1000*rtS/2

This will plug the scan value of rtS in and then calculate the value of p1 and p2 before running the CalcHEP.

Now you can ask your question or file the problem using
https://answers.launchpad.net/hepmdb
available at the front page of HEPMDB

Spheno3.1.4 and softsusy-3.3.2 packages are installed at HEPMDB
and are ready to be used.

The Softsusy is installed at $HOME/softsusy , so to use Softsusy with, for example, CalcHEP models one should add
-L$HOME/softsusy/lib/ -lsoft
line to your extlibN.mdl file.

Analogously, the SPheno package is installed at $HOME/spheno , so to use SPheno with, for example, CalcHEP models one should add
-L$HOME/spheno/lib/ -lSPheno
line to your extlibN.mdl file.

P.S. All news about HEPMDB like this one will be sent to all users registered at HEPMDB (they also have an option to opt out receiving these news)

We suffered a failure of the Iridis cooling system earlier this morning, which led to temperatures in the data centre rising very rapidly. We were forced to take down most of the Iridis nodes to protect the Univerities IT infrastructure. Temperatures in the data-centre are now under control again. We are led to believe the cooling should be fixed soon and we will then be able to start bringing back nodes. However, we do not expect to be able o resume a batch service until after lunch. Apologies for the disruption to your work.

"CalcHEP and HEPMDB: practical introduction and tutorial"
will take place at 15:00 CERN time, Friday, May 04
in the room
304-1-007
is the subject of final of the final confirmation

The link to indico page with the respective materials is at

https://indico.cern.ch/conferenceDisplay.py?confId=189668

Tutorial will be broadcasted in EVO at
http://evo.caltech.edu/evoNext/koala.jnlp?meeting=MtM8Ma2t2aDuDu9D92Ds9t

LHAPDF is installed at HEPMDB and can be used now.
To use LHAPDF installed at HEPMDB with CalcHEP models one should add

-L$HOME/lhapdf/lib/ -lLHAPDF

line to your extlibN.mdl file.

P.S.
All news about HEPMDB like this one will be sent to all users registered at HEPMDB
(they also should have an option not to receive these news if they want)

At this workshop will be presented the first stage of development of the HEPMDB which is already a convenient centralized storage
environment for HEP models, and can accommodate, via web interface to the HPC cluster, the validation of models, evaluation of LHC predictions and event generation-simulation chain. The ultimate goal of HEPMDB is perform an effective LHC data interpretation isolating the most successful theory for explaining the LHC observations.



The schedule and registration are available at http://indico.cern.ch/event/hepmdb

A new database is available to download from our system. It is possible to validate this model on our system as well. A new database is available to download from our system. It is possible to validate this model on our system as well.

We now have confirmation from IBM that all the Iridis 3 racks will be arriving on the 26th September. 

The current Iridis 3 is our third generation cluster and is one of the largest computational facilities in the UK.