Jim,
After I got my DEC spike problem solved with my 811 (the key adjustment was to reset the tension on my SLW boxes, as per a Scott Losmandy video that Brian sent me - the adjustments eliminated the backlash and loosened the box so that it would float as designed - it was over tightened and frozen) I went back to AP adventures. Last month I wanted to try and put together a mosaic of the Cygnus Loop. I spent seven sessions gathering data for the four panels with my ASI2600 MC Pro. I am using NINA for acquisition and PI and a trial license with APP for mosaic processing. ?As I got into the processing, I found that my astro files (9 months of data and processing on about 35 DSO's) had maxed out my 2 TB SSD drive. I had a storage problem.
I began to read and watch videos and talk to a few nearby friends. It seemed I was in the market for a NAS. Most of the advice suggested that I get a unit that could handle five years of data collection as I look to the future. That made sense. They also suggested the 3-2-1 strategy, 3 copies of your data on 2 different media with 1 offsite. I thought that was probably a bit of overkill for me. A third piece of advice had to do with read/write speeds of hard drives and transfer speeds between my computer in the NAS unit. ?
I purchased a Synology DS218+ double bay unit, but swapped it out for a DS940+ 4 bay unit when I discovered the 218+ could not accept future expansion units. The setup was simple enough even for someone like me without any networking experience. I installed 2 10TB hard drives (leaving 2 bays available for future expansion), plugged it into my router, managed to set up a storage pool, a couple volumes, some folders, and transferred my Astro data over. I also ran an ethernet 6 cable from my router to my workstation. It all seems to work quite well, and I am pleased with it at this point.
I have had to make one important adjustment in my processing workflow. The Cygnus Loop mosaic project has been massive, with 1400 X 60s subs, gathered over seven sessions and stacked into four different panels. As you know, Jim, our ASI2600s produce 52 MB subs! Then there were the layers of folders to get from the subs, to integrated masters, through the assembly process.?I ended up with 28 masters for my panels that were 940 MB each, before I ever got to my a final master. ?When I tried to handle this workflow with my computer hardwired to my NAS unit, the data transfer was too slow. So what I ended up doing was transferring all the subs to my newly empty SSD drive and plugging it directly into my computer until my processing was completed. That resolved any transfer delay issues. After I am done, I will plug the SSD into the NAS UBS3 port and transfer all the data back to the NAS. Easy to do.?
I will be very interested to know how you resolve the data storage issue.
I am also very interested to know what data of people keep long term. I assume that I will be adding fresh data to my DSO folders in the future and will want to process everything again. I also assume that the software is developing so rapidly that reprocessing my data with the new software competencies will sooner or later be worth doing. So what should I keep? I would really prefer not to have to keep everything. Do I keep the raw fits files with the calibration data? I assume that I keep my masters. I noted that some people said they kept only their registered folders plus their masters. But if you wanted to preserve your ability to re-process, wouldn't you have to keep something earlier, like perhaps the calibration folders? I'd be glad for any advice.
Geoff Chapman
Chip Louie
Jamie Amendolagine