Data Uploading Advice
Posted: Sat Oct 07, 2017 3:26 pm
Hey Folks,
I recently picked up a new Meteobridge PRO on backorder and I'm pretty excited to get it working with a remote weather station (Davis Vantage Pro2 + LTE Gateway) that I have running.
In the meantime, I'm trying to understand more about what my data uploading options are.
Since the Meteobridge is going to be powered by a solar panel and battery, I've already disabled the wifi in order to conserve battery life. Once it's operational, I'll disable the screen as well. Limiting the number of running push services will probably also help. (Aside: other recommendations?)
What I would like to understand is: "What would be the best approach to upload all possible weather data that the Meteobridge collects in near real time?" My plan is to limit most of the processing on the Meteobridge itself and do most of the data crunching and rendering in the cloud.
It's a lofty ask and I realize that there will be some limitations -- with that said, here is what I'm curious about:
* Is it better to upload the data via FTP or HTTP? Are there limitations on how much data can be passed from Meteobridge via HTTP? Are there any known performance issues/considerations when using (S)FTP?
* This maybe more appropriate for FTP -- is there a comprehensive template that the community recommends using? Obviously I can define my own and do file/string parsing in the cloud, but I didn't know if people here have experiences/preferences on using a dense file format (for smaller uploads) versus a common file format so that it could be later consumed easily by other systems or people. I'd like to upload everything that I can now, so that in the future I'll have the historical information I need when I realize that "I should have been uploading this all along"
* Recommendations on how often to upload the data? i.e. "I wouldn't do anything less than 1 minute, the Meteobridge will do X"
Although at some point the weather data will be stored in a database, I'm thinking that using the Meteobridge mysql support might be too intensive for my system -- hence why I'm exploring FTP/HTTP as possible solutions.
Anything else that I should consider thinking about/doing?
Thanks!
I recently picked up a new Meteobridge PRO on backorder and I'm pretty excited to get it working with a remote weather station (Davis Vantage Pro2 + LTE Gateway) that I have running.
In the meantime, I'm trying to understand more about what my data uploading options are.
Since the Meteobridge is going to be powered by a solar panel and battery, I've already disabled the wifi in order to conserve battery life. Once it's operational, I'll disable the screen as well. Limiting the number of running push services will probably also help. (Aside: other recommendations?)
What I would like to understand is: "What would be the best approach to upload all possible weather data that the Meteobridge collects in near real time?" My plan is to limit most of the processing on the Meteobridge itself and do most of the data crunching and rendering in the cloud.
It's a lofty ask and I realize that there will be some limitations -- with that said, here is what I'm curious about:
* Is it better to upload the data via FTP or HTTP? Are there limitations on how much data can be passed from Meteobridge via HTTP? Are there any known performance issues/considerations when using (S)FTP?
* This maybe more appropriate for FTP -- is there a comprehensive template that the community recommends using? Obviously I can define my own and do file/string parsing in the cloud, but I didn't know if people here have experiences/preferences on using a dense file format (for smaller uploads) versus a common file format so that it could be later consumed easily by other systems or people. I'd like to upload everything that I can now, so that in the future I'll have the historical information I need when I realize that "I should have been uploading this all along"
* Recommendations on how often to upload the data? i.e. "I wouldn't do anything less than 1 minute, the Meteobridge will do X"
Although at some point the weather data will be stored in a database, I'm thinking that using the Meteobridge mysql support might be too intensive for my system -- hence why I'm exploring FTP/HTTP as possible solutions.
Anything else that I should consider thinking about/doing?
Thanks!