Kerberos Factory

Bring state-of-the-art scale and high availability to your video surveillance or analytics landscape

Kerberos Factory
High availbility and fail over. We make sure you sleep well.
Build for scale
Supporting over thousands of IP cameras.
Enterprise Support
Surround yourself by experts in video surveillance.

Built for Enterprise needs

A solution that scales with your expectations

Kerberos Factory lifts the concept of Kerberos Agent to the next level. With deep integration into Kubernetes, it allows you to scale beyond the horizon. It breathes high availability, fail over and delivers a never seen scalability.

Kerberos Enterprise Agent

Video surveillance on Kubernetes

Kerberos Agents as deployments in your K8S cluster

A simple UI build on top of the Kubernetes API, allows you to schedule Kerberos Agents inside your Kubernetes cluster with just a couple of clicks. Updating and upgrading your agents is all done within a few seconds.

Video surveillance on Kubernetes

Kubernetes CLI and Graphana

Leveraging the entire toolset of Kubernetes

As Kerberos Factory is build on top of Kubernetes, it works 100% with the entire Kubernetes ecosystem and toolset.

You can use the tools you already master, such as the Kubernetes command line, prometheus grafana dashboards, nginx ingresses and much more.

IP Camera agnostic

Connect your RTSP and H.264 supported cameras

We do not rely on a set of IP camera vendors, we bring them all! No vendor lock in. Kerberos Enterprise adapts to your landscape. Kerberos Enterprise works with any H.264 supported RTSP stream.

IP Camera agnostic

Scalability, high availability and fail over

We got your back, all industry best practices in a single platform

No longer worry about scaling your video surveillance landscape. Kerberos Enterprise is build on top of Kubernetes and allows you to attach the number of nodes your video landscape requires.
High availability
Connect multiple nodes to your clusters, and leverage the concept of Kubernetes deployments. Kerberos Enterprise makes sure your cameras are always on in any circumstance.
Fail over
If one of your nodes go offline, another node will take over the Kerberos agents. When one of your cameras go down you will be notified, and Kerberos will try to reasteblish a connection until it’s available.


Per instance per month

Cluster installation € 0 / month

Ready to give Kerberos Factory a spin?

Download now and get up and running in a few minutes

Extend and integrate for most features

Supercharge and scale your experience with any of these Kerberos products

Kerberos Vault
  • On-premise and cloud storage
  • Offline capabilities
  • Extend with API’s and Message Brokers
  • Machine learning integration
Kerberos Hub
  • Camera monitoring and inventory
  • Recording and livestreaming
  • Grouping and fine grained access
  • Tasks and notifications FAQ

Frequently asked questions

Questions & answers

hub Where is my data hosted if i'm using the Kerberos Hub public edition?

When connecting one or of your cameras to the Kerberos Hub, your recordings will be stored in a GCP datacenter, Cloud Storage, in Europe (St. Ghislain, Belgium).

In future releases it will be possible to select a region of interest.

enterprise Is Kerberos Enterprise licensed?
No Kerberos Enterprise is not a licensed product, you can install it free of charge, and start experimenting with it into your surveillance landscape.
Is Kerberos only limited to video surveillance?

No Kerberos Enterprise is used in many video analytics scenarios, where it acts as a scalable video platform. Kerberos Enterprise is deployed in different sites, is monitoring the different cameras and persisting recordings.

After having those recordings are stored, those are analysed in real-time by video analytics solutions and algorithms. Kerberos Enterprise helps in having a scalable and error prone video landscape backbone.

Can I install the Kerberos Enterprise suite in my own datacenter / at the edge / on premise?

Yes, that’s one of our design principles, every solution we are and will develop in the future can be self-hosted wherever you want.

We are shipping Docker images for all our solutions, and recommend to use Kubernetes to deploy and maintain them.

vault Kerberos Vault is priced per cluster what does this mean?

Kerberos Vault is installed in a Kubernetes cluster, and for every additional cluster you are setting up Kerberos Vault, you will need an additional license.

The reason why you might need additional Kerberos Vault’s is because of edge processing. You might want to store your recordings locally, and process them. After detecting a hit, you can forward the recording to a centrally hosted Kerberos Vault.

Open source Can I use Kerberos Open source completely free of charge?
Yes, of course! We want to give back to the community, and we will keep working and improving this.
Roadmap Can I influence the development of the services?
Yes you can, both our Open Source and Enterprise suite products are continuiously extended with new features. You can check the roadmap for more details.
Hub Can I install Kerberos Hub in my own datacenter and whitelabel it?

Yes, Kerberos Hub can be installed anywhere where you want. We have a helm chart available for you to install the entire Kerberos Hub solution.

Please note that you will need to request a license to be able to download the appropriate docker images.

Hub Can I integrate my own machine learning models into Kerberos Hub?

Yes, next to storing your recordings with Kerberos Vault you can also use it to integrate with custom machine learning pipelines.

Imagine, you have cameras making recordings and persisting them in your preferred storage with Kerberos Vault. Everytime a recording comes in, Kerberos Vault will fire an event (Kafka, SQS, etc) which can be consumed by a subscribed machine learning pipeline.

After the machine learning pipeline did its magic, it will have generated some metadata that then can be forwarded in the Kerberos Hub queue.

Hub What is the Kerberos Hub Queue about?

The Kerberos Hub comes with a bunch of microservices for pre and post processing the recordings. To streamline these microservices we have implemented Kerberos Hub queue.

This queue, which relies on Kafka, allows our customers to scale, have high availability in place and most importantly implement custom code and functions.

You could perfectly add a post processing, call an API, or whatever you have in mind. Extensions and both integrations to any system are possible.

Hub Can I update the Kerberos Hub layout?

The Kerberos Hub is closed source, so you will only received the binary and Docker image when purchasing a license. However we might include or co-develop Angular components or features together and include that in the Kerberos Hub UI.

Features might also be scheduled in our roadmap, which are then and planned and executed depending on the customer requests and priorities.

Enterprise/Open source Which cameras do you support?

With the Kerberos Open source agent we are supporting: USB cameras, Raspberry Pi cameras and IP cameras (MJPEG, RTSP/H264). To make this work Kerberos Open source will do encoding on the CPU and GPU if running on the Raspberry Pi (broadcom drivers).

For Kerberos Enterprise agents we only support RTSP/H264 encoded streams. Thanks to this we do not have to do any encoding and we can build a more scalable solution.

Hub How is Kerberos Hub licensed?
The Kerberos Hub is a fixed monthly fee independent from the number of cameras of sites. You pay per installation on our private cloud, or get an unlimited license for on premise deployment.
Vault Can store recordings on my own storage?
Yes, that’s the reason why we have build Kerberos Vault in the first place. We want our customers to persist their recordings wherever they would like them.
Enterprise Is Kerberos Enterprise suite licensed per camera?

The agent is completely free of charge and can be installed without license. So you don’t have to pay for any camera.

A license is required when you start storing recordings in Kerberos Vault, and/or want to visualise and manage your recordings, cameras and sites through Kerberos Hub.

Vault Can I have Kerberos Vault in multiple regions to improve latency?

Yes, 100%. Within a single Kerberos Vault you can create multiple providers. Each of those providers (AWS S3, GCP, MinIO, etc), might be stored in a different region.

Depending on where the agents are located, and the end users monitoring the recordings or ml pipelines evaluating them, you might need to store them in a closer region.

With Kerberos Hub you can assign a specific site to a Kerberos Vault, so the best latency can be guaranteed.

Enterprise/Open source What is the difference between Open Source and Enterprise?

The answer is scale. Kerberos Open Source was initially developed to be a low cost, DIY, video management system accessible for everyone. Kerberos Open Source works great and our community loves it, but it doesn’t bring the scale you are looking for.

Due to this the Kerberos Enterprise suite was developed to overcome scale, but also allow customer to store their recordings wherever they want, and have a scalable UI to manage, monitor and scale their surveillance landscape.

Hub Does has livestreaming capabilities?
Yes livestreams are visualised in the Kerberos Hub solution. When using Kerberos Open Source you will see a low resolution stream (JPEG), when using the Kerberos Enterprise agent you will have high resolution streaming using WebRTC (on top).
Enterprise Does Kerberos Enterprise run on K3S?
Yes, it can both run on Kubernetes (K8S) and on K3S. The installation might be slightly different as in K3S you already have the Traefik ingress build in.
Sorry, no questions found

Try searching for a different phrase or use the topics filter to narrow your request. If you don’t find what you’re looking for..

Contact us

Need assistance?

Browse Docs or reach out to our community or Enterprise support

All the features and guides in one place
Community Support
Reach out to our Reddit community
Enterprise Support
We’re here 24/7 for your needs