How Microservices Benefit Broadcasters

了解更多关于微服务的信息，请参见流媒体的下一个事件.

Read the complete transcript of this clip:

马克Myslinski: 现在, 如果你能实现微服务, 你可以找到集装箱, you can get to that which is optimized through the internet and cloud topology and IP-based protocols. 现在 we can get away from the dedicated stuff. 现在我们来看看云拓扑. 现在 we have the expandability if we need that for volumes of processing or volumes of traffic or whatever. I see that as some big benefits and those applications become cloud-native, 而不是仅仅启用云. They can run on a cloud, a cloud-native. And the Googles and the Facebooks and the Amazons of the world, 他们是吃这些东西长大的, 所以它们已经在那里了. And if the broadcasters are gonna try to compete with them they've gotta get to this cloud-native kind of applications. I think that's one of the big benefits once they can get there, that they can truly compete with agility and scalability of the big guys that I just mentioned, 但他们必须这么做.

保罗MacDougall: I think there's also some big advantages around, 你知道, the scalability. 经常, you might get a 新 initiative that we need to put content on some other platform or something like that, and the time frame of that might be something where you have to scale up a lot of resources, especially if you do that with a monolithic architecture, 你知道, you have to scale up aspects of your application that aren't necessarily beneficial to that process. So, 你也许还能再吃点, old-school transcoder and put that on a VM and, 你知道, 启动一千个虚拟机来做这件事. 但很有可能, 您正在复制服务, you're replicating software and components and so forth that are using resources that aren't beneficial to, 你知道, 整个项目. 这就是效率损失. 有迭代的能力, 当然, 新版本, testing out just the component that you need that 新 feature on. 它非常强大, 特别是我们的市场, 你知道, constantly releasing 新 HDR formats, 新的编解码器, 新, 你知道, all kinds of 新 bells and whistles all the time. And then the last one that I think is a big trend that we've seen a lot lately is the isolation of responsibilities between different departments. 所以有些东西, especially larger broadcasters have done, they have the IT department that single responsibility becomes now just installing servers, even if they're not going the cloud-native route, they're going and just racking generic CPUs in some data center somewhere. And the fact that they can be abstracted from what software's running out of the considerations that might be unique to that software, because it's running on some sort of microservice, some sort of container architecture, means that they don't need to get into the nuances of what that application requires.