That’s a good question! We did a lot of research way back when to try and pick the best protocol with the best security. We really liked COAP for being a W3C standard, and for its similarity to HTTP / other common internet protocols.
We’ve also found that most MQTT implementations these days tend to be insecure by default. From what I’ve seen, many of the secure 3rd party implementations for that protocol are also really memory and resource heavy, making it difficult to run them on many devices. It’s definitely something we could be open to if there were enough demand. I’d totally be open to some kind of MQTT client library that would let you push data to a 3rd party while using the primary secure connection to manage firmware and your device, etc, etc.
@Dave So what are most of your larger clients doing to integrate with these enterprise level database services?
It seems AWS and Azure all default to a TSL based MQTT data transfer method as their default method of communication.
AWS and Azure are considered one of the largest database platforms out there, aren’t they?
Is the only downside to using the Webhook method to get data into Azure the extra and processing needed to get it into the database? With the biggest downside being an additional cost due to the extra processing service required to push the data from the webhook into the Azure database?
Sorry for all the questions I’m just a beginner trying to figure out the easiest path to dumping data into Azure.
Is there a recommended tutorial for doing this based on today latest firmware and advancements?
Would you say this is what I should be trying to follow to get going?
AWS / Azure / Google are big infrastructure providers that make it easy to host databases without having to buy dedicated server hardware. Paul’s example you linked is a good way to get started, and the secure database connector will make it easier still.
We see a lot of customers who already have an existing data solution or ingress point, so things like custom database connectors or middleware apps go a long way. Typically these can include a transformation or alerting action as well, so having custom software catching that live data can often be really helpful.
I’ll probably update my hook / azure example to demonstrate how you could import data directly into a database on Azure without using event hubs or stream processing, just so I can make hosting for that demo cheaper.
That would be awesome, and that's exactly what I'm hoping to accomplish.
From what I have seen Particle + Azure + Microsoft Power BI = one of the best online data storage and dashboard solutions out there. The Power BI software suite for visualizing the data created by Particle devices along with other data sources gives you ultimate flexibility in displaying that data how you want.
Power BI provides mobile and tablet apps that allow you to view your Power BI dashboards on Windows, Android, & iOS devices including Apple Watches which is killer.
How hard to you think it would be to modify your code to import data directly into a database on Azure? I'd be willing to throw up some cash if that helps get it done.
I would like to create a solution possibly using Azure + Power BI and submit it on the Hackster.io challenge they have going on right now.
I would hate to set this up using an inferior method to what you're planning on cooking up. Let me know either way.
After watching videos of the Power BI platform over another day, I can say that it’s one of the best solutions out there for creating, sharing and visualizing all this IOT data.
I think Particle should look info partnering with Power BI to provide this solution to its ecosystem if they are considering heading in this direction.
I have heard of PubNub, but I’m going to look into it more now. It looks like it could be a decent all in one solution for what I’m doing possibly, especially since they integrated Power BI into it.
I did find a PubNub article on how to get a Photon publishing to PubNub which is a good start!
And then once you have the data stream setup in PubNub you can use this toutorial to get that PubNub stream to show live data in a Power BI dashboard:
