Computing at BaPSF
BaPSF provides several computing resources for users while at UCLA, in addition to, publicly available IDL and Python
packages for accessing data from the generated HDF5 files. While at UCLA, the following resources are available:
- High speed wireless connectivity
- Data analysis and visualization software developed by the local group
- Workstations for data processing and making computer generated videos of acquired data
High-speed wireless networks are available in STRB and the facility offices, which is connected via a 10 Gbit uplink to the main UCLA campus. The STRB also has a dedicated 10/100/1000 Mbit switch with a direct fiber optic link to the facility office space next door.
A Linux server hosts 148 TB of RAID5 storage for collected. BaPSF also keeps a local backup of collected data on external hard drives.
Two workstation are available for users. Midas.pa.ucla.edu has 72 cores and 512 GB RAM. Ethanol.pa.ucla.edu has 16 cores and 48 GB RAM. The work stations are equipped with IDL, Matlab and several language compilers (C, C++, Python, Fortran, etc.). Terminals are available in the LAPD control room to connect to the work stations. One Mac is available the 3D visualization suite AutoCAD Maya. Users can request a user-account from the UCLA Physics department computing help desk for access to midas and ethanol. These accounts may be used while users are at their home institutions for the purpose of accessing and analyzing data acquired at BAPSF using the facility servers.
A 3D HDTV with stereo shutter glasses is available to view rendered images and movies of complex plasma phenomena.
In addition to local resources, BaPSF has developed software packages in IDL and Python to assist with reading HDF5 files and analyzing data.
Using data collected at BaPSF
Python: BaPSF has developed the python package bapsflib to assist users with managing and analyzing collected data. The bapsflib package is developed for Python 3.5+ and is registered with PyPI (https://pypi.org/project/bapsflib/). The package can be easily installed using pip or installed/downloaded from GitHub (https://github.com/BaPSF/bapsflib). Documentation is hosted on Read the Docs (https://bapsflib.readthedocs.io/en/latest/).
The package is currently focused on providing a high-level, structured user interface to access data stored in the HDF5 files. To do this, bapsflib leverages the well-developed h5py and numpy packages.
In the future, the package will add expanded LaPD tools (calculating field profiles, parameter regimes for LaPD setups, etc.) and analysis tools (fitting Langmuir sweeps, integrating b-dot signals, etc.).
IDL: BaPSF has standard routines available to read in the HDF5 data files into IDL. The suite of routines is available for download. The routines allow for easy importing of data, probe positions as well as machine state information (magnetic field configuration, discharge current/voltage, interferometer array data.