Jump to: navigation, search


What is HEPMDB?

HEPMDB is a database that stores high energy physics models, and calculates theoretical cross sections for user defined processes in order to validate theory against experimental data. There are other projects similar to this, however HEPMDB is the only database that allows users to upload models, run processes and calculate cross sections on the site itself.

The model database can be accessed by everyone. However, in order to use HEPMDB's computational capabilities you will have to register first.

Current Capabilities

  • Users can upload model files and model source files (i.e LanHEP, FeynRules) for various Matrix Element Generators inc. CalcHEP, MADGRAPH, WHIZARD etc.
  • HEP simulations can be performed using any of the public models (currently 50+) uploaded to the site by registered users, or by your own privately uploaded model.
  • Calculations are performed in the HPCx cluster interface, providing up to 16 cores per user. This allows for faster calculation and larger jobs than on a desktop or laptop.
  • Model history is stored so that any modifications to the model itself can be tracked.
  • Kinematical distributions of various properties can be plotted and stored in the Les Houches Event (.lhe) file format.
  • Can search for and download existing models from HEPMDB.
  • Option to leave reviews for other's models to aid development.

Future Prospects

The database is still in development, here are some of the possible prospects for the future development of HEPMDB.

  • Addition of the LanHEP and FeynRules packages, and the CompHEP Matrix Element Generator.
  • Allow users to upload their own Feynman rule or Matrix Element Generator packages.
  • Compare BSM predictions with LHC data systematically in order to discriminate between the different theories.
  • Extend database structure to deal with correlated signatures.
  • Make format for model predictions consistent with LHC data format used by experimentalists.
  • Install MicrOMEGAs package to evaluate dark matter relic density, and to provide scans of model parameter spaces.
  • Possibly study events beyond parton level, up to detector simulation.

Getting started

How do I register on HEPMDB?

To register, all you need are basic details such as your name, email and affiliation. Your account will be reviewed and approved by one of our administrators. You will be notified by email once your account is approved. Registration can be done on the HEPMDB main site. As usual, all fields marked with an asterisk are mandatory. Check for further help here.

Keeping Updated

Now and again the programs used on HEPMDB (CalcHEP, Whizard, MadGraph) will be updated to their newer versions. When this happens, to use the newest version it is necessary for the user to reset their account via the menu at the top right of HEPMDB's calculate section. Warning!: All files and models will be deleted from your account!

General usage

What is the Calculate section?

This section allows you to use the models uploaded on HEPMDB and directly compute results using HEPMDB computers. Only once registered will you be able to use HEPMDB Calculate section. More information on how to use this section is provided here.

Which programs can I use on HEPMDB?

Currently, three different programs can be used for computation on HEPMDB: CalcHEP, WHIZARD and MadGraph. Any model developed for these programs and uploaded on the HEPMDB website will be accessible.

Why should I use HEPMDB's Calculate section?

There are many practical reasons for using the HEPMDB computing mainframe. Convenience may be the most obvious one; no installation is needed and it is simple to compare results between various matrix element generators. Also, HEPMDB has some integrated tools for plotting and analysing that are really quick to use.

Another reason is simply the efficiency; with 16 cores, HEPMDB is much faster than most local installations.

Where can I find some help on how to use the Calculate section?

These wiki pages will provide help. Here you can find a specific help page for the Calculate section, with information on the interface and descriptions of the different available programs: CalcHEP, WIZARD and MadGraph.

Becoming an author

Who is an author?

If you want to use the HEPMDB computation abilities, you will have to become an author. Authors are also granted the right to upload new models (and encouraged to do so). In the future, they will also be able to add model signatures.

How do you upload a model?

As soon as your account is approved, you can upload new models. This is done through the upload model page.

Further help can found here.

The batch file

How can I assign one parameter to another?

In general, you cannot assign the value of one parameter to another in the symbolic form. The parameter will be set to zero and may cause a crash.

Instead, you can run a loop over some fake parameter, setting it to be the desired value. For example:

Run parameter: new

Run begin: 2000

Run step size: 500

Run n steps: 1

Then you can assign both parameters to this fake parameter:

Parameter: lc=new

Parameter: MUU=new

How do I exclude a process in the batch file?

To remove intermediate particles you should use the "Remove" statement in the Process Info section.

For example, for the process p,p->A,A you could exclude the Higgs particle to find its background as follows:

Process: p,p->A,A

Composite: p=u,U,d,D,s,S,c,C,b,B,G

Remove: h

This will remove any Feynman diagrams that involves Higgs interaction such as G,G->A,A and one diagram of b,B->A,A.

How do I scan over the collider energy?

This can be done by defining a fake parameter eg: rts in the run parameter section:

Run parameter: rts

Run begin: 7

Run step size: 1

Run n steps: 2

Then in the momentum info section, write some function of this fake parameter:

p1: 1000*rts

p2: 1000*rts

This will change the collider energy per run that the file scans over by this user defined fake parameter 'rts'.

How do I keep the change in mass between particles constant when running over a particle's mass?

You can define a fake parameter to run over which you then add to your starting mass. For example, if we wanted to change masses M1 and M2 but keep the gap between them at 50 GeV:

Parameter: M1=50+dM

Parameter: M2=100+dM

Run parameter: dM

Run begin: 10

Run step size: 10

Run n steps: 20

It's important to remember that you cannot use symbols when defining a parameter unless it is a fake parameter from the run parameter section.

Using the results

Why can't I download the results lhe file?

It should be noted that it is not possible to download files bigger than 1GB from HEPMDB. Unfortunately, file sizes are not indicated yet. The consequence of this download limit is that you cannot combine too many lhe files into one before uploading.

Making a model for HEPMDB

How do I use an SLHA input file?

If you wrote a model, it may be using an SLHA input file. On HEPMDB, there is an option for the user to upload his/her own SLHA file. In order to use that one, you should use the file ../models/SLHA.txt in your model. Any SLHA file that the user uploads will be renamed and relocated accordingly.

Presentations on HEPMDB

Miniworkshop on High Energy Physics Model Database (HEPMDB)

March 2 2012, Durham, IPPP [1]

CalcHEP and HEPMDB: practical introduction and tutorial

by Alexander Belyaev, Friday, 4 May 2012, CERN [2]

HEPMDB: practical introduction and tutorial

Alexander Belyaev, 16 August 2013, CERN [3]