Can we keep our Grafana dashboards?

Yes. InfluxDB is a supported target, so existing dashboards keep working while Proxus takes over collection and modelling underneath.

Which protocols do the open-source components cover?

Modbus, OPC UA, MQTT and Siemens S7. Most other vendor PLC, CNC and robot protocols are not covered; Proxus has native drivers for 66 protocols across 30 vendor and standard families.

What does Proxus cost compared with a free stack?

A licence priced by gateways and data rate, from $5,500 a year as published, with the maintenance inside it. The open-source stack has no licence; its cost is engineering time and maintenance for as long as it runs.

Comparison · A DIY stack

You can build this. The question is who maintains it in year three.

Telegraf, InfluxDB and Grafana (with a few scripts and maybe Node-RED) will collect industrial data and chart it, for the price of the hardware. Plenty of good plants run exactly this. It is a real option and we will not pretend otherwise; the honest comparison is about what happens after it works.

Book a demo All comparisons

What it is: A DIY stack: Open-source assembly, typically Telegraf, InfluxDB, Grafana and scripts
What Proxus is: An industrial data platform: collect, model, govern, store, deliver
Honest summary: Cheap to start, genuinely capable, and the maintenance is the product you end up owning.
Your options: Usually one or the other: keep building, or move collection, modelling and history into Proxus and keep the parts that earn their keep, such as Grafana. Both paths are below.

Give them their due

What A DIY stack is genuinely good at.

A comparison that only lists the other product's weaknesses is a sales sheet, not a decision aid. These are real strengths, and some of them are reasons not to buy Proxus.

Excellent components, freely available

Telegraf ships as a standalone binary with hundreds of plugins including MQTT, Modbus and OPC UA, and buffers at the edge. InfluxDB and Grafana are mature. None of this is a compromise.
No licence conversation

You can start this afternoon without procurement, and prove value before anyone asks for a budget line.
Complete control

Every part is inspectable and replaceable. For teams with the skills and the appetite, that is worth more than any vendor relationship.
Skills that transfer

What your team learns building it applies far beyond this project, which is not true of most vendor-specific tooling.

Attribute by attribute

Where the two actually differ.

Both columns are filled in. Where A DIY stack is ahead, the row says so.

Time to first chart

Proxus

Comparable for the first device, and the modelling work is what takes the extra time.

A DIY stack

Fast. An afternoon for a familiar protocol.
Industrial protocols

Proxus

66 protocols across 30 vendor and standard families natively, including CNC and robot controllers.
In the docs: Industrial protocols

A DIY stack

Modbus, OPC UA, MQTT and Siemens S7 are covered. Most other vendor PLC, CNC and robot protocols are not, and that is where the custom work starts.
Data model

Proxus

A modelled namespace with units, quality and asset context, with changes recorded in the audit trail.
In the docs: Data model

A DIY stack

Measurement and tag names agreed by convention. The model lives in dashboards, queries and people.
Access control

Proxus

One role model across data, screens, APIs and AI clients, including namespace paths.
In the docs: Access control

A DIY stack

Per component: database users, Grafana roles, host access. Coherent per-signal permissions are hard to assemble.
Configuration management

Proxus

Pipelines keep revisions you can compare; every other configuration change is recorded in the audit trail.
In the docs: Configuration management

A DIY stack

Config files and dashboards in version control if the team is disciplined about it.
Where configuration lives

Proxus

Centrally on the hub. A replacement gateway is installed with a new command from the hub and pulls the same devices, rules and functions.
In the docs: Where configuration lives

A DIY stack

Config files on each host, in version control if the team is disciplined about it.
Multi-site

Proxus

Templates and a gateway fleet with central visibility.
In the docs: Multi-site

A DIY stack

The same stack per site, kept consistent by process. Drift is the normal failure mode.
Moving parts at a site

Proxus

One gateway process plus a messaging container for the local buffer, installed by one script. The seams between components are inside one codebase rather than between four.
In the docs: Moving parts at a site

A DIY stack

An agent, a database, a dashboard server and usually a broker: four products to version, patch and supervise.
Who supports it

Proxus

A vendor with release notes, certified processes and a support channel.
Certificates: Who supports it

A DIY stack

Your team, plus community channels. The person who built it is the escalation path.
Cost shape

Proxus

A licence priced by gateways and data rate, from $5,500 a year as published, with the maintenance inside it.
Pricing: Cost shape

A DIY stack

No licence. Engineering time up front and, more importantly, maintenance for as long as it runs.

Attribute	A DIY stack	Proxus
Time to first chart	Fast. An afternoon for a familiar protocol.	Comparable for the first device, and the modelling work is what takes the extra time.
Industrial protocols	Modbus, OPC UA, MQTT and Siemens S7 are covered. Most other vendor PLC, CNC and robot protocols are not, and that is where the custom work starts.	66 protocols across 30 vendor and standard families natively, including CNC and robot controllers. In the docs: Industrial protocols
Data model	Measurement and tag names agreed by convention. The model lives in dashboards, queries and people.	A modelled namespace with units, quality and asset context, with changes recorded in the audit trail. In the docs: Data model
Access control	Per component: database users, Grafana roles, host access. Coherent per-signal permissions are hard to assemble.	One role model across data, screens, APIs and AI clients, including namespace paths. In the docs: Access control
Configuration management	Config files and dashboards in version control if the team is disciplined about it.	Pipelines keep revisions you can compare; every other configuration change is recorded in the audit trail. In the docs: Configuration management
Where configuration lives	Config files on each host, in version control if the team is disciplined about it.	Centrally on the hub. A replacement gateway is installed with a new command from the hub and pulls the same devices, rules and functions. In the docs: Where configuration lives
Multi-site	The same stack per site, kept consistent by process. Drift is the normal failure mode.	Templates and a gateway fleet with central visibility. In the docs: Multi-site
Moving parts at a site	An agent, a database, a dashboard server and usually a broker: four products to version, patch and supervise.	One gateway process plus a messaging container for the local buffer, installed by one script. The seams between components are inside one codebase rather than between four. In the docs: Moving parts at a site
Who supports it	Your team, plus community channels. The person who built it is the escalation path.	A vendor with release notes, certified processes and a support channel. Certificates: Who supports it
Cost shape	No licence. Engineering time up front and, more importantly, maintenance for as long as it runs.	A licence priced by gateways and data rate, from $5,500 a year as published, with the maintenance inside it. Pricing: Cost shape

Proxus figures come from the current build's own driver and target inventory. A DIY stack figures come from the vendor's published material, linked at the end of this page. Nothing here is measured on a benchmark neither party has agreed to.

Choose A DIY stack

When we are not the answer.

One site, a handful of signals, and an engineer who wants to own it: a platform would be overhead.
The protocols involved are Modbus, OPC UA or MQTT, which the open components already handle well.
The team has genuine capacity for the maintenance, not just the enthusiasm for the build.
The data is for engineering curiosity rather than decisions the business will defend in an audit.

Choose Proxus

When the shape of the problem fits.

The stack became load-bearing and the person who built it has moved on, or would like to.
The equipment includes controllers the open plugins do not speak.
A security review needs one access model, not four, and an audit trail behind the numbers.
Site two and site three have to look like site one without depending on the same engineer.

If this column describes you:

Run Proxus next to your stack: request an evaluation key

Not either/or

How they are used together.

Keep Grafana, change what feeds it

InfluxDB is a supported target. Existing dashboards keep working while the collection and modelling underneath become someone’s product rather than someone’s script.
InfluxDB integration
Telegraf where it is already good

For infrastructure and IT metrics, Telegraf stays the better tool. Proxus does not need to own everything that produces a number.
Keep the scripts that earn their keep

Genuinely custom logic can move into a C# function inside the pipeline, where it is governed and supported rather than scheduled by cron.

Sources

Last checked against vendor material: October 2026.

Products change. If something on this page is out of date or wrong, tell us and we will fix it, including when the correction favours A DIY stack.

Questions

Proxus and A DIY stack: short answers.

Can we keep our Grafana dashboards?: Yes. InfluxDB is a supported target, so existing dashboards keep working while Proxus takes over collection and modelling underneath.
Which protocols do the open-source components cover?: Modbus, OPC UA, MQTT and Siemens S7. Most other vendor PLC, CNC and robot protocols are not covered; Proxus has native drivers for 66 protocols across 30 vendor and standard families.
What does Proxus cost compared with a free stack?: A licence priced by gateways and data rate, from $5,500 a year as published, with the maintenance inside it. The open-source stack has no licence; its cost is engineering time and maintenance for as long as it runs.

Proven in production

Running in production at these sites.

All deployments

Other comparisons

All comparisons

Next step

Compare year three, not day one.

Install Proxus next to what you built, point it at the same controllers and compare the work it takes. An evaluation key gets you started; an engineer answers the questions it raises.

Request an evaluation key Book a demo

01
Scoping call

Equipment, protocols, sites and the systems the data has to reach.
02
Technical walkthrough

A working demo against a scenario that resembles your plant, not a generic tour.
03
Pilot plan

A defined first line or site, with the deployment and security questions answered up front.
04
Roll-out

The same templates bring the next lines and sites online.

You can build this. The question is who maintains it in year three.

What A DIY stack is genuinely good at.

Excellent components, freely available

No licence conversation

Complete control

Skills that transfer

Where the two actually differ.

Time to first chart

Industrial protocols

Data model

Access control

Configuration management

Where configuration lives

Multi-site

Moving parts at a site

Who supports it

Cost shape

When we are not the answer.

When the shape of the problem fits.

How they are used together.

Keep Grafana, change what feeds it

Telegraf where it is already good

Keep the scripts that earn their keep

Proxus and A DIY stack: short answers.

Running in production at these sites.

Compare year three, not day one.

Scoping call

Technical walkthrough

Pilot plan

Roll-out