coder . cl

As we have seen Wowza Media Server supports custom extensions through its Java API. One big deal with the Wowza Media Server, while I was creating the time shifted streaming extension, was the server thread synchronization. Not a big deal from the point of view regarding the client side stream, but the monitoring thread for each recorder and playlist was a problem to synchronize. Most problems were related to interface implementations.

Wowza Media Server supports a wide variety of applications thanks to its extensible API. The media server API is exposed as a set of Java classes and interfaces to be implemented, so you can handle server side, application side and stream side events, metadata and media control routines. I’ve implemented a stream recorder because the live-record stream type records all streams on the configured application and also it do not have any configuration directive to control the recorded stream.

As we have seen, the Wowza Media Server is an extensible Streaming Server that can allow you to create time shifted streams or scheduled streams. If you are using the live-record stream type or the LiveStreamRecord module, you can create a good time shifting module to schedule streams along a given set of streams. The problem is that the playlist must be dynamic, so I’ve solved that issue rescheduling the playlist.

Wowza Media Server is a Tomcat server adapted to serve audio and video streams using the RTMP protocol. One great advantage of this Media Server is the fact that you can extend it using Java modules. The API is a little bit hard to understand, but very flexible. One of my recent projects was to create a Dynamic Scheduler Module for the Wowza Media Server.

Java synchronization is usually made through the synchronized keyword. Synchronized allow users to create a mutex around certain variable, class or method, where a mutex allows concurrent access to it. By concurrent access we should understand access from multiple threads. If an operation is atomic, there one and only one process or thread executing it. Then, mutex based operations are atomic.