My plan is to create a new file every day (for the next three years...). Into this file will go the datasets for that day (around 24 Gbits). For the file (aka container file) I would like to distinguish between science, calibration and housekeeping data. The science (and possibly the calibration file) would also be distinguished by instrument name. To do this I was thinking of calling the files: eis_sci_yymmdd eis_cal_yymmdd eis_hk_yymmdd xrt_sci_yymmdd etc... where yymmdd is the two digit year, month and day. In this scheme the level-0 and level-2 fits datasets would appear in the same file. If required these coukld go into different files. So straight away there are two questions to ask: Is it possible to name files like this? What is the largest allocation I can give to a file? (On a related note what happens if I then try to write more data to the file than I have allocated?) The naming scheme for the datasets in the eis_sci_yymmdd file is: eis_sci_yymmdd_hhnnss.fits where yymmdd is as above and hhnnss is hour, minute, second (generated from the spacecraft time), and for eis_cal_yymmdd (calibration files): eis_drk_yymmdd_hhnnss.fits (dark current) eis_flt_yymmdd_hhnnss.fits (flat field)