-->
Save your seat for Streaming Media NYC this May. Register Now!

视频:云服务中专用硬件加速的案例

Learn more about cloud encoding and delivery at the next Content Delivery Summit.

Watch Brent Yates's presentation, 改变流交付的经济性:使用专用硬件加速的全栈云服务,在… Streaming Media Conference Video Portal.

Read the complete transcript from this video:

Brent Yates: 我们不可能不先谈软件就谈硬件, 尼尔在这方面做得很好:软件开发的方式正在发生变化. 每个人都在使用云,但云不是一个物理的地方. 这只是做软件和构建这些微服务的一种方式, of designing for the problem, not for the hardware you're running the problem on.

软件设计师希望能够构建一个软件, and define what the requirements are, and then have that run on whichever server it happens to be, without having to know what OS is on it, how much RAM is on it, how many drives are on it, what processor is on it. Those are secondary concerns. What they care about is solving the business problem. 这看起来很简单,它应该是这样的, 但我们花了很长时间才达到完美的状态,拥有了能够构建这种系统的所有功能.

Now with these automated orchestration layers, and languages, 和工具, you can do that, you can focus on solving the problems, 让自动规则把你的软件推向边缘. 在这种情况下,在边缘上提供的服务是CDN的一部分. It gives you a couple of good features. 它更容易管理,因为你关注的是问题,而不是硬件. 您可以拥有异构服务器部署,其中它们具有不同的硬件需求.

But the programmer doesn't have to know that. 软件会自动找出在哪里以最有效的方式运行它. 它使企业能够专注于问题并快速适应不断变化的关注点,而不必专注于硬件部署.

Just to reiterate, the cloud can be public cloud like Neil said, 或者是混血儿, or in this case, a unique way of expanding the cloud out to edge nodes. 通过将软件从大型单片应用程序中分离出来,它充分利用了这些高cpu核数, 我指着尼尔展示的第二张幻灯片, 他在哪里有一个大的单片应用程序,然后分解成微服务. 这些轻量级的微应用程序可以更容易地适应您的业务逻辑, and it also means that there's a lot more of them.

这是一个开销,在你的服务器上有数千个微服务会产生摩擦. 但是,通过这种方式分解应用程序所获得的效率抵消了这种摩擦.

那么,为什么要在这种类型的部署中添加专门的硬件呢? 为什么不把它放在常规的英特尔72核硬件上,让它运行呢? 原因是cpu的增长不是基于摩尔定律. Moore's law is dead. We're only getting about 3.5% improvement year over year on CPU performance. So you throw these things out there in your container, you're throwing more services, more problems, more chatty apps, 但唯一能让你变得更快的方法就是等上10年cpu的更新, or throw more servers at the problem.

Throwing more servers at the problem is not only capital-intensive; it's also wasteful. It's not green. It's a power problem. 微软有能力在数据中心上花费90亿美元, and Google can spend $10 billion dollars. But those are big capital deployments. It would be much better if we can make efficiency changes, 并改变我们做生意的方式,而不是仅仅向它扔服务器.

Specialized hardware can do that.

We look at it as a capacity gap problem. 在底部,蓝线是CPU性能,红线是需求. Pretty much everyone we talked to, and we've been going around the room today, 他们要恢复的河流数量是否会以每年30-35%的速度增长, how many apps they have to run, how much data they're putting out. 解决这个缺口的唯一方法就是用服务器来填补它.

Streaming Covers
免费的
for qualified subscribers
订阅 Now Current Issue 过去的问题
Related Articles

是什么 True Cloud and How Is It Changing Streaming?

signant首席营销官Jon Finegold在2019年流媒体西部的演讲中讨论了signant真正的云框架和新兴原生云世界的好处.

Video: What FPGAs Are and Why They Matter

HellaStorm首席技术官Brent Yates在2019年内容交付峰会上的演讲中讨论了现场可编程门阵列(fpga)以及它们如何为云加速提供灵活性.

视频:实时应用需要一个新的动态交付平台

Limelight Networks的尼尔·格莱泽布鲁克在2019年内容交付峰会上的演讲片段中,阐述了向新的动态平台转移以交付实时应用程序的理由.

Video: How Real-Time Apps Feed Our Need for Immediacy

Limelight Networks的尼尔·格莱泽布鲁克在2019年内容交付峰会的这段视频中讨论了实时应用在日常生活中的主导地位.

视频:动态编解码器管理和新兴的多编解码器世界

Akamai首席架构师Peter Chave预测了2022年的多编解码器前景,并解释了动态编解码器管理将如何帮助OTT提供商在2019年内容交付峰会上进行管理.

Video: How Much Bandwidth Will You Need to Deliver 8K?

Akamai首席架构师Peter Chave在2019年内容交付峰会上的主题演讲中讨论了8K流不可避免(但不必要)到来所带来的带宽需求.

Video: 是什么 Women in Streaming Media?

Fastly's Alicia Pritchett and Datazoom's Diane Strutner, co-founders of Women in Streaming Media, discuss the organization's mission, 计划, and opportunities at Content Delivery Summit 2019.

Video: How to Synchronize Streaming Playback

Akamai首席架构师Peter Chave在2019年内容交付峰会上的开场主题演讲中提出了三个步骤来解决流媒体在不同网络上的不同玩家所特有的漂移问题.