I continue to tweak my scripts and data collection on my NMEA 2000 server that I added to last year in Data collection from the sailboat. They are not even close to being ready for prime time, but the data I’m collecting is starting to force me to adapt the system to match the resolutions required. Things like battery voltages are easy because they shouldn’t change every second, but things like heading, heel angle, and other sailing related items should have much more granular resolution.
RRD’s are great for things that have 30 second to 5 minute resolutions, but terrible for things like real time data like the above. I have yet to find any great way to store this data and present it usably other than a static webpage that a client can update at predefined intervals. I’m terrible at some of the newer technologies, but it looks like there are more flexible technologies that will push second-by-second updates to a client.
One of the other things I’ve done is to push differing types of traffic from the bus into a database and start looking at frequencies and organizing things in different ways. There are some obvious consistencies and items that are repeated over and over, which never would change – devices claiming their addresses and such that you wouldn’t want to log all of the time.
There are definitely a lot of challenges in developing something like this for the commercial market. Maretron and others have N2K monitors and PC’s that parse lots of this data and will take actions, but no one seems to have a full boat monitoring or data collection system yet…
Below is a copy of the current network on Jammy. Quite a lot of endpoints and interfaces, but I love working on her! Click for a larger version and zoom in.