Abstract
The Hall-D Online Skim System (HOSS) was developed to simultaneously solve two issues for the high intensity GlueX experiment. One was to parallelize the writing of raw data files to disk in order to improve bandwidth. The other was to distribute the raw data across multiple compute nodes in order to produce calibration skims of the data online. The highly configurable system employs RDMA, RAM disks, and zeroMQ driven by Python to simultaneously store and process the full high intensity GlueX data stream.
Highlights
The GlueX Experiment[1] is a high volume Nuclear Physics experiment installed in experimental Hall-D of the Jefferson Lab Continuous Electron Beam Accelerator Facility
The Hall-D Online Skim System (HOSS) was built to simultaneously address a need to produce skim files online while reducing I/O bandwidth to individual RAID servers as needed by high intensity running of the GlueX experiment
The system uses Python programs applying rules defined in a configuration file making it a portable system that could be applied to other experiments
Summary
The GlueX Experiment[1] is a high volume Nuclear Physics experiment installed in experimental Hall-D of the Jefferson Lab Continuous Electron Beam Accelerator Facility (aka CEBAF). The raw data files would need to be distributed among several RAID servers in order to reduce the average rate any one server needed to support Another issue that came up while processing the phase I data was that it took considerable effort to extract special calibration events from the stored data files. The DAQ system implementation for GlueX could not be changed to write separate output streams for these events directly They needed to be extracted from the full raw data set. An ability to generate these skim files in the counting house before the raw data ever made it to tape would save considerable time and effort in the offline Implementing this ability in the existing DAQ system would have been difficult so it was done using a separate, new system called HOSS for the Hall-D Online Skim System. These components as well as details of HOSS itself are given
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
