谷歌挑战AWS如何

大公司经常被批评为“错过”了未来——从舒适的栖息的一份礼物,表示未来,当然,但在未来仍然是未来在职者往往可能最好的例子是微软: the company didn’t “miss mobile” — Windows Mobile came out in 2000 — but rather was handicapped by its allegiance to its license-based modular business model and inability to envision a world where its core product (Windows) was a planet orbiting mobile’s sun; everything about Windows Mobile’s design presumed the exact opposite.

One could make the same argument about Google and the enterprise; both G Suite (née Google Apps for Your Domain) and Google Docs launched a decade ago and enjoyed modest success, particularly in smaller businesses and education; unsurprisingly, both markets share broadly similar characteristics to Google’s core consumer user base — limited configurability and a low price were good things牵引被大企业更难获得,不过,事实上在过去的几年中Office 365很好的节奏G套件,不仅增长快,赢回顾客。

Still, for all the success Microsoft has had with Office 365, the real giant of cloud computing — which is to say the future of enterprise computing — is, as is so often the case, a company no one saw coming: the same year Google decided to take on Microsoft亚马逊推出了亚马逊网络服务AWS之所以如此引人注目的方式,它反映了亚马逊本身:它是建立在规模和明确和硬接口客户-首先亚马逊还公司在世界各地访问“原语”,可以混合和匹配建立一个更有效率,可扩展和安全端比几乎任何公司可以建造的。

AWS的原语

今年早些时候在亚马逊的税收我解释了Amazon AWS策略源自相同的方法使公司成功的:

该公司由多个相对独立的团队组成,每个团队都有自己的损益,责任和分布式决策[所有的商店作者布拉德]石头解释早期贝佐斯倡议(强调我的):

The entire company, he said, would restructure itself around what he called “two-pizza teams.” Employees would be organized into autonomous groups of fewer than ten people — small enough that, when working late, the team members could be fed with two pizza pies这些团队将独立释放在亚马逊最大的问题……贝佐斯是一种混沌理论应用于管理,承认他的组织的复杂性把它分解到最基本的部分希望令人惊讶的结果可能会出现。

Stone later writes that two-pizza teams didn’t ultimately make sense everywhere, but as he noted in后续的一篇文章该公司仍然非常广泛分布的平面与责任在那里,在那些“最基本的”,是借自己的原语规模和实验记住上面的引用描述贝佐斯和团队抵达AWS的想法:

If Amazon wanted to stimulate creativity among its developers, it shouldn’t try to guess what kind of services they might want; such guesses would be based on patterns of the past相反,它应该创建原语——计算的基石——然后走出。

Steven Sinofsky喜欢指出组织倾向于船组织结构图,当我开始显示亚马逊复制AWS模式,事实证明,AWS模型在许多方面的表示亚马逊本身(就像iPhone在许多方面反映了苹果的单一组织):创建一个堆原语,让开,一个好的脱脂。

AWS的提供无疑扩大了远远超出基础设施(虚拟化)处理器一样,硬盘驱动器,和数据库,无论是进一步抽象(如λ“serverless”计算)和栈分成平台和软件服务,但它的成功的基础仍是亚马逊的纯平台的方法:他们为企业提供部分建立任何他们想要的。

谷歌是一个产品的公司

Google, meanwhile, has never really been a platform company; in fact, while Google is often cast as Apple’s opposite — the latter is called a product company, and the former a services one — that only makes sense if you presume that only hardware can be a product更广泛的“产品”定义 - 向最终用户展示的完全实现的解决方案 - 将显示两家公司实际上非常相似。

别搞错:云服务和硬件之间的差异是深远的(我探索的长度苹果公司的组织的十字路口),但公司的产品之间的差异和一个平台理想的产品,无论是智能手机还是一个搜索框,实现简单性和一个伟大的用户体验设计和工程通过巨大的努力,理想情况下,由最终用户从未见过确实,这就是为什么集成产品在消费市场中取胜,毫无疑问,谷歌以消费者为中心的服务历来是集成后端iphone。

但是,请注意,这是完全相反的模型受雇于不仅亚马逊也微软,杰出的平台公司IT时代: instead of integrating pieces to deliver a product AWS went in the opposite direction, breaking down all of the pieces that go into building back-end services into fully modular parts; Microsoft did the same with its Win32 API是的,这意味着Windows按设计更大平台的终端用户体验,说,Mac操作系统,但它是更强大的和可扩展的,这种方法了数以百万计的业务线应用程序即使在今天把Windows的核心业务AWS为后端服务做了完全相同的事情,AWS的灵活性和模块性是其粉碎谷歌最初的云产品Google App Engine的主要原因,该产品于2008年推出Using App Engine entailed accepting a lot of decisions that Google made on your behalf; AWS let you build exactly what you needed.

谷歌的平台解药

窗户的例子是有益的在思考如何谷歌已经改变了方法:大规模生态系统建立在微软的广泛的API最终最终锁定最明显的是,为Windows构建的应用程序并不容易移植到其他操作系统,但同样重要的是庞大的合作伙伴网络和增值经销商使Windows成为企业唯一可行的选择亚马逊正在努力构建相同的生态系统。

然而,它从来没有更可行使用Windows,首先对消费者也为企业,原因是网络:这是一个新的运行时坐在窗口的顶部,但不依赖于它,1在消费者方面,谷歌是最大的赢家事实上,浏览器解释AWS的崛起:任何新的业务应用程序构建web(包括应用程序运行在基于web api),它是在任何设备访问。

事实证明,在过去的几年谷歌浏览器进行一种企业计算方法。在2014年谷歌宣布Kubernetes,开源容器集群管理器基于谷歌内部Borg服务抽象谷歌,谷歌这样的大规模基础设施服务可以即时访问所有他们所需要的计算能力,而不用担心细节中央规则是容器,其中2014年我写了:工程师建立在一个标准接口,保留(几乎)完整的灵活性,而不需要了解底层硬件或操作系统(在本是一个进化步骤超出虚拟机)。

Kubernetes与Borg的不同之处是它是完全可移植:它运行在AWS,它运行在Azure,它运行在谷歌的云计算平台,它运行在本地基础设施,你甚至可以运行它在你的房子吗More relevantly to this article, it is the perfect antidote to AWS’ ten year head-start in infrastructure-as-a-service: while Google has made great strides in its own infrastructure offerings, the potential impact of Kubernetes specifically and container-based development broadly is to make irrelevant which infrastructure provider you use难怪这是增长最快的一个开源项目的:没有锁定。

But how does that help Google? After all, even if Kubernetes becomes the standard for enterprise clouds Amazon’s broader ecosystem lock-in is still present (and the company has its own container strategy that further locks customers into AWS); Google needs a differentiator.

成本和经验

Here again the desktop is instructive: the open nature of the web running on platform-agnostic browsers did not make Google successful per se; rather, the openness of the web created the conditions for the best technology to win和谷歌不仅有最好的搜索引擎,但它之所以是最好的——的依赖关系,而不是简单的页面内容,意味着随着网络变大,谷歌与竞争对手不同的是,有更好的。

I think this is an idea that can be abstracted to be broadly applicable; indeed, it’s a core piece of聚合理论:随着分布(或开关)成本降低,增加用户体验的重要性换句话说,当你可以访问任何服务,不管是新闻或拼车或酒店或视频或搜索等,一个是最好的将不仅赢得最初将其优势化合物。

This is Google’s bet when it comes to the enterprise cloud: open-sourcing Kubernetes was Google’s attempt to effectively build a browser on top of cloud infrastructure and thus decrease switching costs; the company’s equivalent of Google Search will be machine learning.

机器学习和数据

似乎可以肯定,机器学习将越来越多地由云服务:都是关于处理规模和大量的数据,而且只有少数巨头有财务能力不仅构建所需的基础设施,也有必要采用世界上最好的机器学习工程师通过扩展,意味着对大多数企业分化引起的机器学习将首先来自他们的数据是否在云中(会有内部的解决方案,但我期望他们会随着时间的推移越来越多的背后),其次从云提供商他们选择哪一个。

That raises the stakes for cloud providers themselves; superior machine learning offerings can not only be a differentiator but a sustainable one: being better will attract more customers and thus more data, and data is the fuel by which machine learning improvement comes about这是因为谷歌的数据AWS云的最大威胁。

我描述了谷歌的企业业务有限的消费上,但谷歌的巨大的优势是它一直在与大量的近20年来的数据,和发展中强大的机器学习算法在过去几年不过,重要的数据最重要的是,和最好的证据是这样去年谷歌开源TensorFlow,机器学习的蓝图:正如我所指出的TensorFlow和货币化的知识产权谷歌愿意分享其方法隐晦地承认其优越的数据和处理基础设施是一个可持续的优势。

我们刚刚开始看到这种优势适用于Google的云产品就在感恩节谷歌之前一系列的产品发布显然,利用其数据的优势:

  • 云自然语言API,它使用机器学习来分析文本,毕业一般可用性
  • 高级版的云翻译API,它使用机器学习大大提高翻译的准确性八种语言(超出标准版,支持超过100种语言)
  • 大降价云愿景的API,它使用机器学习来分析图像
  • 一个新的云工作API,使用机器学习与工作匹配潜在雇员

这四个加入了Cloud Prediction API,它使用机器学习来做出预测It, along with the first three APIs above, is clearly derived from various Google consumer products; the Jobs API likely builds on an internal Google tool, as well as Google’s wealth of data from all over the web在每种情况下,谷歌花了数年时间磨练其算法的应用于企业数据集的结果很有可能优越,或者至少漏斗的培训我希望这个优势仍然存在,是有意义的。

Still, Google will have to do more, which is why the other big announcement was the creation of the Google Cloud Machine Learning group headed by Fei-Fei Li and Jia Li: this group will be charged with building new machine learning APIs specifically for business; to put it another way, they are tasked with productizing Google’s machine learning capabilities.

That, in a roundabout way, gets to the genius of Google’s strategy: the company was outpaced by Amazon in the first wave of cloud computing because success rested on being the best platform; by open-sourcing Kubernetes in an attempt to shift the industry to vendor-agnostic containers, Google is trying to move the plane of competition to products毕竟,常常很容易改变规则的竞争比改变你的本质作为一个公司。


可以肯定的是,谷歌的成功并不保证:公司还必须应对一个新的商业模式,销售和广告,建立必要的组织不仅销售企业的支持在这两个领域,亚马逊有一个开端,一个大大大的合作伙伴生态系统和一个更大的特性集。

当然,AWS机器学习有自己的API,以及IBM和微软Microsoft is likely to prove particularly formidable in this regard: not only has the company engaged in years of research, but the company also has experience productizing technology for business specifically; Google’s longstanding consumer focus may at times be a handicap和流行Kubernetes可能是广泛的,它是关于谷歌尚未吃自己的狗粮.

不过,谷歌将是一个强大的竞争对手:其策略是合理的,或许更重要的是,找到一个新的业务的紧迫性更紧迫的今天比2006年的水平Most importantly, the shift to cloud computing is still in its beginning stages, and while Amazon seems to be living the furthest in the future, the future has not happened yet; it will be fascinating to watch Google’s attempt to change the rules under which said future will operate.

  1. 尽管有ActiveX [↩︎]