FRB Observations

FRBs are high transient by their nature. As a result, we could see a single peak in a 20 hour observation. While keeping the raw voltages on hand can help with re-processing observations to avoid missing features due to issues in the current methodology, we no longer have the space to perform such actions.

Process the observation at CDMT. Produce both a 0-DM and N-DM filterbank at the nominal time resolution (currently 327us, 8x ts, 8x chan)
Search this output after an 8-bit decimation from digifil
- Log the heimdall commands used, RFI channels flagged
- Investigate the optimal scale timescale (-I) for FRBs; some are expected to last up-to or over 1 second at our frequencies due to scattering. Digifil default is definitely lower than this.
After a search is complete, archive the CDMT filterbanks
- Digifil: 2x space saving compared to raw filterbanks, output
  - -I 0 : No scale changes
  - -b-32 : Float32 output, no change from raw filterbank
  - -t 2 : Down sample to 655us resolution
- Compressed with zstandard: Further 20-40% compound storage saving
Future changes
- Reduce bandwidth? Top 5MHz / Bottom 10 are noise + RFI contaminated
  - Removing these could save us 15% of storage and speed up processing as searching the last 10MHz introduces an addition delay of 25 seconds @ R3's DM
  - No easy way to do this with the current voltage extraction/processing method, would need to be after the filterbanks are formed

Step	Method	Storage Used	Product	Overall on Disk
Generate Voltages	Observer	1	1
Compressed	zstandard	~0.6-0.8	0.6	0.6
CDMT	-a -b 8 -d 0,DM,2	0.25	0.25	0.85
Digifil (Search)	-b 8 -I <DECIDE>	0.0625	0.0625	0.9125
Cleanup: Digifil (search)	rm	-0.0625	-0.0625	0.85
Digifil (Compress)	-b-32 -t 2 -I <DECIDE>	0.125	0.125	0.975
Cleanup: Voltages, CDMT	rm	-0.6 - 0.25	-0.85	0.125
Overall				~140GB/hr