OpenVidu Enterprise



OpenVidu
CE
OpenVidu
PRO
OpenVidu
ENTERPRISE
WebRTC media streams
Recording
Screen sharing
Audio and video filters
IP cameras
Clients: JS, Angular, React, Vue.js, Ionic, Electron, React Native
Deployment on Premises
Deployment on Cloud
OpenVidu Inspector
Advanced session monitoring
Manual scalability
Elasticity
S3 recording storage
Network quality API
Virtual Background
Fault Tolerance
Media Server Kurento Kurento Kurento / mediasoup
Streams per core 100 100 500
High Availability
Simulcast
VP9
E2E encryption
Large scale sessions
License Apache 2.0 Commercial Commercial
Support Community support on forums Email support in business hours Email support in business hours
Price Free 0.0006$ core/minute
0.0018$ core/minute
Docs Docs Docs

OpenVidu Enterprise features 🔗

OpenVidu Enterprise offers the best performance, improved media quality and better scalability for high-demand environments.

The key feature of OpenVidu Enterprise is that it supports using mediasoup as Media Server instead of Kurento. mediasoup brings a lot of benefits for videoconferencing apps built with OpenVidu:

5x more media streams 🔗

Using the same hardware, OpenVidu Enterprise with mediasoup supports up to 5 times more media streams than with Kurento.

5x quicker connections 🔗

Media connections are established 80% quicker using OpenVidu Enterprise with mediasoup. This means that with Kurento the average time between a customer calling the subscription operation and the video being actually played on their device, it could average around 1.25 seconds. With mediasoup it averages 0.25 seconds.

Media quality improvements 🔗

OpenVidu Enterprise with mediasoup raises the bar of what is possible with adaptive video quality for WebRTC, compared to what was possible with Kurento:

  • Simulcast

    Simulcast is a technique that allows optimizing the quality of routed video, in accordance with the needs of each individual Subscriber. Depending on aspects such as device form-factor or network link quality, the video that gets delivered to each participant in a session can be adjusted to be the perfect match for each circumstance.

    This comes in contrast with the more traditional method of adaptive video bitrate in Kurento, which was all-or-nothing and affected all participants equally, so it wasn't possible to adjust the quality individually for each one of them.

    For more technical details about how simulcast works, check the Simulcast technical details page.

  • VP9

    An advanced video codec that brings better quality features when compared to VP8 and H264.

    VP9 achieves better compression rates and better quality with lower bandwidth usage. It also supports SVC, which is in itself an improvement over the benefits provided by Simulcast.

    Popular web browsers such as Google Chrome have integrated VP9 support for a while now, other browsers are slowly getting onboard, and OpenVidu will allow you to benefit from it too.

High Availability deployment option 🔗

OpenVidu Enterprise offers a High Availability deployment option in AWS, with replication of all nodes and load balancing for clients. Visit the High Availability documentation for further information.

100% compatible with your current OpenVidu applications 🔗

OpenVidu hides all complexity that lies behind swapping Media Server technologies: New SDKs, architecture changes, etc...

With OpenVidu, there's no need to change a single line of your application: what used to work with OpenVidu Pro will work as-is with OpenVidu Enterprise.



Kurento vs mediasoup 🔗

OpenVidu Enterprise offers the possibility of choosing between two different media servers for routing media:

  • Kurento is a powerful full-featured media server used in OpenVidu CE and OpenVidu Pro. It is based on low-level processing of the media streams. This provides advanced features such as transcoding and real time audio-video filters. There is a downside to this power: it can bring some overhead for videoconferencing applications that do not need these advanced features.

  • mediasoup is a modern, lightweight and fast media server. Its performance for routing media in typical videoconferencing applications is higher than Kurento, as mediasoup does not process media streams at a low level like Kurento does. Besides, it brings the latest technologies in media transmission, such us simulcast, VP9 media codec and E2E encryption with Insertable Streams.

In general, most applications built with OpenVidu will greatly benefit of using mediasoup. The only cases in which it is better to use Kurento is for those applications that specifically require some of the functionality only offered by Kurento (audio/video filters or IP cameras).

Kurento mediasoup
Average number of streams per core (higher is better) 100 500
Average time for media establishment (lower is better) 1.25 s 0.25 s
Audio and video filters (Doc)
IP cameras (Doc)
Simulcast (Doc)
VP9
E2E encryption

Additional mediasoup limitations 🔗

There are a couple of small limitations when using mediasoup that must be taken into account:

Using Kurento in OpenVidu Enterprise 🔗

OpenVidu Enterprise uses mediasoup as default media server. But you can use Kurento instead by overriding configuration property OPENVIDU_ENTERPRISE_MEDIA_SERVER.



OpenVidu Enterprise roadmap 🔗

E2E encryption 🔗

Thanks to mediasoup, OpenVidu Enterprise will offer E2E encryption using WebRTC Insertable Streams. With Kurento, media streams are encrypted in the client-to-server and server-to-client channels, protecting them from man-in-the-middle attacks. But media streams have to be individually decoded and processed in the server side, so data must be decrypted by Kurento, which breaks the client-to-client encryption. But with mediasoup, media streams can remain protected client-to-client, without the server needing to decrypt it.

Large scale sessions 🔗

OpenVidu Enterprise will support much larger sessions in terms of users and streams. This is a statement based on 3 points:

  • The better performance of mediasoup allows processing more media streams in the same hardware.
  • Sessions will be able to be replicated in different Media Nodes, sharing the load of the same session in different machines. This will provide horizontal scaling in OpenVidu for the first time.
  • Selection of dominant speaker(s) will add the possibility of sessions with hundreds or thousands of publishers without crashing client devices. Only the latest active speakers in a session will be sent to the client side, theoretically allowing for an unlimited number of publishers in the same session.



Deploying OpenVidu Enterprise 🔗


You can deploy OpenVidu Enterprise:


Visit Pricing section to learn more about the cost of OpenVidu Enterprise.