===============================
BEATS Computing Infrastructure
===============================
DAQ workstation - ``BEATS-control-ws``
--------------------------------------
The data acquisition (DAQ) workstation is used to control the beamline and scan settings. Visit the section :doc:`daq` for more information.
Useful commands
~~~~~~~~~~~~~~~
.. highlight:: bash
:linenothreshold: 1
Mount ``PETRA``::
sudo mount -t nfs 10.1.14.100:/PETRA/SED/BEATS/IH /PETRA/SED/BEATS/IH
SMB mount of SSCAN data folder::
sudo mount -t cifs -o vers=3,username=beats.smb '\\10.1.14.100\pco-flir-ws' /home/control/Desktop/SSCAN
Start the `Energy GUI `_::
cd /home/control/energy/iocBoot/iocEnergy_2BM
python3 -i start_energy.py
.. highlight:: none
Data analysis workstation - ``BL-BEATS-WS01``
---------------------------------------------
The data analysis workstation is used for several purposes including:
* Inspection of sinograms and CT reconstruction
* Submit reconstruction jobs on the cluster ``rum@sesame.org.jo``
* 3D image visualization and processing
The list of software available on the workstation is listed in the section on :ref:'Data analysis software' below.
Useful commands
~~~~~~~~~~~~~~~
.. highlight:: bash
:linenothreshold: 1
Start `alrecon `_ CT reconstruction environment::
conda activate tomopy
solara run alrecon.pages --host localhost
Start reconstruction pipeline on Jupyter Lab. Available pipelines are described in section :doc:`reconstruction`::
conda activate tomopy
jupyter lab
Mount proposal ``ExpData`` and ``recon`` folders on BL-BEATS-WS01::
cd ~
./petra_prop_mounter.sh
Mount proposal ``ExpData`` and ``recon`` folders on ``User-BEATS-Dispenser`` and ``Dragonfly Dispenser 2``::
./petra_prop_recon_smb_mounter.sh
Mount proposal ``ExpData`` and ``recon`` folders on ``Dragonfly Dispenser 1`` for remote access::
./petra_prop_recon_smb_mounter_vizserver.sh
Check mount points::
df -h
Unmount proposal folders::
umount /PETRA/SED/BEATS/SEM_6/20235010
umount /PETRA/SED/BEATS/SEM_6_recon/20235010
Mount proposal ``ExpData`` and ``recon`` folders on Win Data Dispenser and Dragonfly VizServer::
./petra_prop_recon_smb_mounter.sh
.. highlight:: none
.. note::
For proposals belonging to a different semester the scripts ``petra_prop_mounter.sh`` and ``petra_prop_recon_smb_mounter.sh`` must be modified.
Data analysis software
----------------------
The software in the table below can be used to inspect and process 3D image data (sinograms and CT reconstructions) at SESAME BEATS.
+-----------+-------------------------------------------------+-------------+------------------------------------------------------------+
| Name | URL | Open source | Features |
+===========+=================================================+=============+============================================================+
| ImageJ | https://fiji.sc/ | yes | Essential for data collection and reconstruction |
+-----------+-------------------------------------------------+-------------+------------------------------------------------------------+
| Paraview | https://www.paraview.org/ | yes | 3D image rendering |
+-----------+-------------------------------------------------+-------------+------------------------------------------------------------+
| Dragonfly | https://www.theobjects.com/dragonfly/index.html | no | 3D image analysis and visualization |
+-----------+-------------------------------------------------+-------------+------------------------------------------------------------+
| 3D Slicer | https://www.slicer.org/ | yes | 3D image analysis and visualization |
+-----------+-------------------------------------------------+-------------+------------------------------------------------------------+
| TomoPy | https://tomopy.readthedocs.io/en/stable/ | yes | CT reconstruction in Python |
+-----------+-------------------------------------------------+-------------+------------------------------------------------------------+
| Alrecon | https://github.com/gianthk/alrecon/tree/master | yes | Web app for CT reconstruction |
+-----------+-------------------------------------------------+-------------+------------------------------------------------------------+
| Jupyter | https://jupyter.org/ | yes | Interface for Python reconstruction pipelines (notebooks) |
+-----------+-------------------------------------------------+-------------+------------------------------------------------------------+
Load reconstructed volume with ImageJ
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Reconstructions at SESAME BEATS are generally saved as a stack of ``.TIFF`` images contained in a reconstruction folder. To load a reconstruction in ImageJ use the command ``File › Import › Image Sequence``. You can follow `this video `_ for a detailed explanation on how to import image sequences.
.. figure:: /img/imagej_image_sequence.png
:align: center
:alt: Import image sequence in ImageJ
.. note::
Always select the option ``Use Virtual Stack`` when you import large image stacks in ImageJ!
rum - BEATS reconstruction cluster
----------------------------------
.. highlight:: bash
Access the reconstruction cluster ``rum@sesame.org.jo`` with::
ssh -X beatsbs@rum.sesame.org.jo
.. highlight:: none
Data dispenser PC - ``User-BEATS-Dispenser``
--------------------------------------------
The data dispencer Windows PC (``User-BEATS-Dispenser``) can be used to access, visualize, and copy your beamtime data. You can also use this PC to launch the ``Dragonfly VizServer 2``, reserved to the beamline users during their shifts.
ORS Dragonfly VizServer
-----------------------
We offer the following 2 seats for image processing using ORS Dragonfly to our users.
+-------------+-------------------------------------+--------+-----------------------+----------------------------------------------------------+
| | CPU | RAM | GPU | Note |
+=============+=====================================+========+=======================+==========================================================+
| VizServer 1 | Intel Xeon Silver 4314 CPU@2.40 GHz | 256 GB | (1×) NVIDIA RTX A6000 | Contact the beamline staff to request access (see below) |
+-------------+-------------------------------------+--------+-----------------------+----------------------------------------------------------+
| VizServer 2 | Intel Xeon Silver 4314 CPU@2.40 GHz | 256 GB | (1×) NVIDIA RTX A6000 | Reserved for users at the beamline |
+-------------+-------------------------------------+--------+-----------------------+----------------------------------------------------------+
BEATS data portal / Dragonfly VizServer 1
-----------------------------------------
**SESAME BEATS data portal:** visit `https://access.sesame.org.jo/beats `_.
On the portal you will find instructions on how to access BEATS data and our Dragonfly VizServer 1.