Knowledge

Table of Content

Table of Content

Table of Content

On Demand Vs Turbo On Demand

Understanding On-Demand Streaming in Xtream AI

Xtream AI Streaming Engine introduces two distinct modes for handling channel activation: On Demand and Turbo On Demand. Both are designed to help administrators optimize bandwidth and resource usage, especially when dealing with large channel libraries or restreamed content. However, each operates differently in terms of performance and resource allocation.

What is On Demand?

The standard On Demand mode allows channels to remain in a hibernated state until a user requests playback. When a client attempts to watch a channel:

  • Xtream AI checks which sources are active.

  • Once a valid source is found, FFmpeg is initiated to begin processing the stream.

  • Only then does the stream become available to the user.

Key Benefits

  • Bandwidth savings: No active processing or streaming takes place while the channel is idle.

  • Connection efficiency: Ideal for restream setups where minimizing unnecessary traffic is important.

Considerations

  • Start-up Delay: Because FFmpeg needs to initialize and process the stream from scratch, startup times can range from 9 to 30 seconds, depending on server load and source responsiveness.

  • Latency: FFmpeg was not designed for ultra-low latency or fast booting of live streams, so delays are expected.

What is Turbo On Demand?

Turbo On Demand is a proprietary feature developed exclusively for Xtream AI to improve the responsiveness of On Demand channels without sacrificing efficiency.

When this option is enabled:

  • The system prioritizes server resources to speed up the channel’s boot process.

  • It reduces the wait time experienced by the client by preprocessing parts of the stream faster than regular On Demand mode.

  • The feature operates only with .m3u8 (HLS) formatted sources, which are natively compatible with this acceleration method.

Key Advantages

  • Faster initialization compared to standard On Demand.

  • No need for constant channel processing—still resource-friendly while idle.

  • Enhances user experience by reducing perceived delays.

Important Notes

  • Not Instant: Turbo On Demand does not provide instant playback, but it significantly reduces the delay.

  • Temporary Resource Spike: It may consume higher CPU or memory briefly during channel startup.

  • Format Dependent: This feature only works with sources using the .m3u8 format. If a restream source is not in HLS format, the system will attempt to locate a .m3u8 variant. If unavailable, Turbo On Demand will be bypassed.

Summary

Feature

On Demand

Turbo On Demand

Idle Resource Usage

Minimal

Minimal

Start-Up Time

8–30 seconds (FFmpeg dependent)

Faster (uses accelerated boot process)

Format Compatibility

Any format supported by FFmpeg

Only .m3u8 (HLS)

Resource Efficiency

High

High (with brief spikes at start)

Ideal Use Case

Bandwidth saving with fewer users

Improved user experience on popular channels

Conclusion:
Both modes are designed to help administrators save bandwidth and reduce server load. Standard On Demand is great for maximizing efficiency, while Turbo On Demand provides a faster, smoother experience for your users, especially for high-demand channels using HLS sources

Xtream AI

·

©

2025

All rights reserved

Xtream AI

·

©

2025

All rights reserved

Xtream AI

·

©

2025

All rights reserved