Derek Tuando, IoT specialist and creator of LoRaDB, talks about why traditional databases often fall short when applied to real-world LoRaWAN deployments, and what changes when data systems are designed with devices—not tables or tags—as the primary organizing principle.

Derek explains what an IoT database actually is, drawing clear distinctions between general-purpose databases, time-series tools, and systems purpose-built for LoRaWAN workloads. He outlines the practical challenges that emerge as projects grow beyond early pilots, including query complexity, usability issues, and the friction teams face when stitching together multiple tools just to visualize and understand device data.

The conversation dives into the core idea behind LoRaDB’s device-first data model, where all data is organized around a device’s identity rather than abstract measurements. Derek walks through how this approach simplifies querying, speeds up exploration, and makes LoRaWAN data more intuitive to work with—especially for small teams, hobbyists, and lean organizations managing thousands to tens of thousands of devices.

Derek also discusses where LoRaDB fits today, including its strengths in ease of setup, open-source accessibility, and built-in visualization, as well as its current limitations around high availability and large-scale enterprise deployments. He shares how the project is being used in production, why it’s designed to complement existing LoRaWAN stacks like ChirpStack, and how future improvements are focused on lowering the barrier for new users rather than chasing complexity.

This episode offers a grounded look at the data layer of LoRaWAN systems, with practical insights for builders, operators, and businesses deciding how to store, query, and actually use the data their devices generate.

Links

Derek on LinkedIn

LoRaDB on Github

• Helium Global IoT Coverage - Want to know if Helium coverage exists where you need it? Check out this map!
• Helium Foundation - The Helium Foundation's IoT Working Group (IOTWG) has generously provided support for the first 6 months of shows, please go check them out and consider using the Helium LoRaWAN as a primary or backup on your next deployment. With over a quarter million gateways deployed worldwide, it's likely that you have and can use Helium coverage.
• Support The Show - If you'd like to support the MetSci Show financially, here's where you can donate on a one-time or an ongoing basis. Thank you!
• MetSci Show - If you'd like to use our IoT or AI Data Value calculators, or you'd like to contact me, the MetSci Show site is the best way to do it.
• MeteoScientific Console - Use LoRaWAN - The MeteoScientific Console allows you to use LoRaWAN today. As long as you have Helium coverage (and you probably do, about 90% of populated areas in the world have a gateway within 2 miles), you can onboard a sensor. You can always check coverage at https://explorer.helium.com and switch to the "IoT" tab in the top right.

Today's guest on MeteoScientific's.

2

The Business of LoRaWAN is Derek Tuando,

an IoT specialist and the creator

3

of LoRaDB, an open source database

built specifically for LoRaWAN devices.

4

In this conversation, Derek explains

what an IoT database actually

5

is, why many general purpose databases

become difficult to use as IoT

6

projects grow, and what makes LoRaWAN data

different from other types of telemetry.

7

We walk through his device first approach

to data modeling how DB is designed

8

to simplify querying and visualization,

and where it best fits today.

9

From hobbyist and small team deployments

10

to real production environments

running tens of thousands of devices.

11

This episode is sponsored

12

by the Helium Foundation and is dedicated

to spreading knowledge about LoRaWAN.

13

If you'd like to try, Helium’s

publicly available global LoRaWAN for free

14

and support

the show, Sign up at metsci.show/console.

15

Now let's dig into the conversation

with Derek Tuando.

16

Derek, welcome to the show man,

thanks for coming on.

17

Yeah, thank you for having me.

How are you, man?

18

I am psyched to have you here.

19

Thank you. I'm doing super well.

20

We recorded the show in late December

and then I managed to botch

21

the audio is what I think happened.

22

So we're rerecording this,

which is fun for both of us,

23

because we had at least one practice

round.

24

Yeah. There we go.

25

Give me some more time to think about

what I'm talking about to you.

26

So tell me.

27

Let's start off with what is LoRa DB?

28

It's an easy to use database

that has device first features in it.

29

Meaning you know that the whole database

is designed with the devices

30

as the most important thing.

31

So everything is organized around

each device,

32

all of the data that goes in there.

33

You know, it's when you query stuff,

it'll be based off of the device's EUI

34

and all the information

will come from that, instead of it

35

being different tags and fields

and everything like influx,

36

it'll be a little bit easier and faster

to query data for each device.

37

Okay, so the the big point or the thing

that the big pain point to solve is like

38

it's faster and easier to query data

than whatever Postgres or influx, etc..

39

Yeah, and easier to when you actually use

40

the UI, it's a little bit

more intuitive than influxes.

41

You click on the device tag

and then you have like tags

42

and measurements

and all these different things

43

that are in place

when you're trying to visualize it.

44

This one is just, you know,

you pick the device that you named within

45

chirp stack.

46

It saves,

it uses that device name and EUI.

47

And then you can query different time

frames and everything through there.

48

And it's just a little bit easier

to use, a little bit more intuitive

49

and probably a bit faster.

50

It can query a pretty big data set. Yeah.

51

And is that why you built it?

52

If you could kind of encapsulate,

I mean, building

53

a whole new system is a pretty big job.

54

Yeah, I built it

55

because of that, because influx was

one of the first ones that I've used.

56

I haven't tried too many others.

57

I think you said you used Cloudflare

I believe.

58

Yeah. Cloudflare is one. Yep.

59

Yeah. Yeah.

60

So there's there's other solutions out

there.

61

This one was

I just built because I wanted something

62

that was intentionally made

for just LoRaWAN devices.

63

It doesn't really ingest

any other things than LoRaWAN devices.

64

So it it makes it a nice

65

addition to all the other applications

that are out there.

66

Got it.

67

And what can people expect

when they use it?

68

So you might need a little bit of command

line knowledge

69

as far as getting it set up.

70

But the documentation is pretty thorough.

71

You know,

I try to make it as simple as possible

72

just by running,

you know, like one bash command.

73

It'll set everything up for you.

74

And then setting up the database,

there's two different things you can do.

75

You can either

just set it up to collect data

76

from all of the devices on chirp stack,

or you can set it up

77

to where you can isolate different

applications into separate databases

78

using the a command line interface

that I created for it,

79

that will help you isolate each database

for each application.

80

That way you can

81

when you go to look at everything,

you're not looking at a thousand devices.

82

You're looking at,

you know, each application's devices.

83

Got it.

84

I mean, it sounds pretty cool

to build a database just for LoRaWAN.

85

What are the weak points

that you think it has where you're like,

86

all right, I've got to shore this up

87

or this is just inherent to the system,

but it's it's worth the trade off.

88

So I would say some weak

89

points would be

there's not a lot of application

90

integration outside of whatever

it's built into this application.

91

So like you couldn't just go to Grafana

92

and start visualizing data

from this database yet.

93

And then if you're trying to use it

at a large scale,

94

it's like,

say you wanted to use high availability.

95

That's something that's not built into it

yet either.

96

So mostly for like the hobbyist

or something,

97

you know, smaller projects,

this is a nice, nice thing to use.

98

It's easier to set up,

and you don't have to know

99

a ton of different applications

just to get started.

100

And then when you say smaller,

what are you talking about?

101

Is that 2000, 5000, 50, 50,000 devices?

102

Like what do we.

103

Yeah.

104

So I don't know about how many devices,

but I do know that you, you know, smaller

105

as in a one man show

or if you, you know, you have just a few,

106

you could probably do a couple tens

of thousands of them of devices.

107

It really just depends on how many people

are working on this at this,

108

you know, helping you out.

109

It's it's really it's really meant

for the smaller

110

like, hobbyist people of, LoRaWAN users.

111

Oh, interesting.

112

And it's would you say

it's more focused on hobbyist or it's is

113

it appropriate for enterprise,

114

or is it something that you think

will grow into an enterprise piece?

115

Yeah.

116

So as far as the enterprise,

it was mainly made for hobbyist,

117

not for enterprise use.

118

You could probably use it

for enterprise use.

119

But like I said, the high availability

isn't there.

120

The querying speed. Yeah.

121

Like you can have 50,000 devices

and it'll log and query really fast.

122

But as far as like I said,

you're going to want high availability

123

probably

for certain enterprise applications.

124

But if you're doing, you know,

you have a smaller company or just,

125

you know, hobbyist, I wouldn't say

just hobbyist, but small companies

126

can get away with using it as well,

but not necessarily huge companies,

127

you know, if they're trying to track it

hundreds of thousands of devices.

128

Maybe, maybe Amazon doesn't

listen to this in general.

129

Right, exactly. Yeah. Yeah. Yeah.

130

I mean but like it's

131

I don't know, they might find good

use for it for internal stuff

132

if they just have, you know,

133

they're watching one warehouse,

they can probably get away

134

with using one at each warehouse.

135

But as far as I availability. No.

136

Is there a difference

between the way that you use it at work,

137

and how the average person who finds

it will use it?

138

Are there differences

in how you can interact with it?

139

Not really.

140

I mean, maybe so I have it set up

141

to where the database is in the cloud

on a on a virtual server.

142

You can run it locally,

but I also run the UI that connects

143

to the database to manage everything

and visualize everything locally.

144

You can also run that in the cloud,

145

but I run that locally

and the other things separate that way.

146

Nobody else really has access to the UI

unless they need it.

147

I guess that's just a security thing

I implemented into it.

148

You could run it in the cloud.

149

It does have authorization and everything,

but it's it's just a little bit.

150

Sometimes people like to keep the UI

off of the cloud

151

and local

hosted to make for security reasons.

152

They don't want everybody looking into

their stuff.

153

Got it.

154

And it's a project on GitHub

so anyone can use it.

155

It's not like you're selling it,

at least not yet. Is that right?

156

Right. Yeah. They can get into it.

They can look at the code.

157

They can change anything they want to.

158

Yeah. And just use it freely.

159

Test it out if they want to.

160

It's honestly I like it

because I'm adding new features in there.

161

At first I was not going to have graphs

and everything built into it.

162

But then I thought like,

why have another piece of software?

163

If I'm trying to make it easier

for these people,

164

why make them find another software

that they can visualize it?

165

I just started adding it in there.

166

The big challenge is going to be

all the devices that are in LoRaWAN.

167

You have to kind of make different

168

since they output different measurements

and different payloads.

169

They we have to kind of make a database

for all this kind of stuff too,

170

where anybody who's using

it can also get access

171

to using any of those devices tools

without having to program

172

it themselves into the software

which which I'm working on right now.

173

Got it.

174

And other devices

that you've put a ton of them on there.

175

And this is working well.

176

This is like,

oh, this is really great for water meters.

177

But I haven't tested out, you know,

solar panel monitors or whatever it is.

178

Yeah.

179

So there's there's quite a few devices

180

that I haven't tested out,

but with the way it saves data, it's

181

honestly it's

just a matter of getting the right widget

182

visualizations in place

because the data is there.

183

And once it gets into the database, it's

just a matter

184

of configuring and setting up widgets

to show in the right way.

185

Because I don't have it like Grafana

and different places like that.

186

You can create a widget

for a measurement in a device.

187

What I'm trying to do here is create

a widget for the whole device,

188

so you can see all the measurements

in one widget.

189

That way, instead of having,

you know, 20 different widgets

190

for a device

that has 20 different measurements.

191

Oh, interesting.

192

So like an 11 and one and or monitor

what you're saying.

193

You want to able to see

all of those things in one.

194

Just all of that information

in one widget,

195

which complicates things a little bit,

but also makes it a lot nicer in the UI,

196

because now you don't have to set up

a different widget for each measurement

197

like you do in Grafana, or toggle

or in any of these other data.

198

Cake.

199

That was one of the big things

that I'm trying to implement in this

200

to make it easier.

201

So you just do it by device

and that's that's

202

where it's getting a little bit tedious.

203

But at the same time, in the end

you only have to do it one time.

204

So right now

I have it saving it into Json file.

205

So anybody can pretty much

take the Json files

206

and create their own widgets

using the templates that I have.

207

It's a little complicated, but okay, once

I have it set up, it'll be easier

208

and easier for the people

209

who are trying to get into it to just,

you know, pick a device.

210

What advice do you have?

211

Add the widget and then visualize

your data and then connect it.

212

And is this something

213

that's going to be pretty easy for ChatGPT

or Gemini or Claude or whatever to.

214

Yeah, exactly.

215

So you can

you can probably just go into Claude

216

and say, you know, create me

a Json based off of it.

217

That's why I have that template there.

218

That way you can just add that template

and it'll create that Json file for you.

219

And then you should just be able

to drag it

220

into the device types folder

and actually create new devices that way.

221

Okay. Pretty straightforward

to create new devices.

222

Are you thinking about how

223

AI in general integrates into this, or

is it more that strategies like, hey, I'm

224

just going to provide clear documentation,

and with that, I can do anything.

225

I could possibly add a way for AI to be

implemented to do it, maybe down the line.

226

Yeah, I could probably add

something like that as a, as an option.

227

You know, it wouldn't be hard to add.

228

You know, somebody adds their API key

229

and then they can just get cloud

or something to generate those,

230

those features, those devices for them,

which might be a pretty good idea to do,

231

do the magic.

232

I mean, it sounds pretty easy

if you've got a template

233

and then you've got a new device,

whether it's a

234

whatever radar, people counter

or whatever, you can just say, hey,

235

here's template, here's device, here's

a documentation from the manufacturer

236

who's documentation from Doug,

go make this thing work.

237

Right, right.

238

And it's even easier than that.

239

Once you add the device to the database,

it outputs the LoRaWAN frame for you.

240

And you can just copy that,

copy that frame

241

and then say, can you pretty much

make this frame work with this template?

242

And then it'll output,

it'll fix the code for you

243

to pretty much make a widget for you.

244

And when did you launch this thing?

245

How long has it been going?

246

So I put it up in GitHub.

247

I want to say late November,

248

but I've been using it personally

for a while now, and

249

like I said, just internally with

within my within the company I work for

250

because we don't have a lot

251

hold ton of devices,

but we're starting to pick up

252

to where it's just easier this way.

253

And for me at least, instead of having

to implement everything into influx.

254

And then I don't know, this is something

that I enjoyed doing for this.

255

This project and it's been working.

256

So I figure, you know, I'm not sure.

257

Yeah, yeah. That's rad.

258

And then do

you have a future set of milestones

259

that you want to accomplish or timeline

laid out like, hey,

260

this is what I want to do

261

as I use it, I find new features

that I want to add to it.

262

There's not really any end goals

besides to make it as easy as possible for

263

for new people to come in.

I think that's that's the main goal.

264

And the more I use it, the more I realize

and the more I use other, other tools.

265

It's like, well, they don't do this

very well or they don't do this yet.

266

And I think it would be a lot easier

if they did it this way.

267

I think it's because Laurel and so new,

268

and there's just not a lot of focus on

just making LoRaWAN tools better.

269

There is on like the application side,

I guess, but the database side is like

270

you choose your own database and

and figure it out.

271

Yeah. Go with that. Yeah, yeah, exactly.

272

And I mean, some places are making it

easier. Data cake.

273

You can ingest everything in there,

but then you don't have a lot of choice

274

with what lenses you use.

275

You kind of just are stuck with theirs.

276

Okay, dude, super cool

to see things being built for LoRaWAN.

277

It's it's been around long enough

that it's

278

nice to see the ecosystem

advancing in this way.

279

Is there anything else that you want

to share or you want people to know about?

280

Lora DB nope, but if I, you know,

if they have any questions,

281

you can reach out on GitHub and I'll try

to answer them as best as possible.

282

And it'd be nice

283

to get people to just start using it

to see what kind of problems we run into.

284

That way we can solve, implement.

285

Yeah, exactly.

286

And see what kind of issues they have

instead of just my issues put into this.

287

Yeah yeah yeah. Well ripping.

Thanks ton for your time.

288

I know, I know you get your regular job.

289

So appreciate you carving out some time

for your day to talk to us about Lora DB

290

yeah, thank you for having me.

291

That's it for

this episode of The Business of LoRaWAN.

292

If you want to go deeper

and actually deploy devices,

293

the MeteoScientific

console is the fastest way to do that.

294

And honestly, it's

also the best way to support the show.

295

When you use the console, you're not just

listening, you're participating

296

in the same real world LoRaWAN work

we talk about here every week.

297

You can get started with the free trial

at meteoscientific.com.

298

Huge thanks to the sponsor of the show,

the Helium Foundation,

299

for supporting open LoRaWAN

infrastructure worldwide.

300

Check them out at helium.foundation

and if the show has been

301

useful give us a quick rating or review

on Apple Podcasts or wherever you listen.

302

This really helps

303

people find it and helps the show grow

so we can help more people.

304

I'm Nik Hawks with MeteoScientific.

305

I'll catch you on the next episode.