Another re:Invent has come and gone, and the dust has started to settle. We talked quite a bit last week about the announcements that were made and the direction Amazon Web Services is taking with these new offerings. But what does it all mean? That’s a complex question with complex answers, so let’s first recap some of the main points from each day.
Monday was a sort of “pre conference” day, but that didn’t prevent AWS from launching some really cool new products and services. The most exciting was Amazon Sumerian, a VR/AR service that looks to make 3D environments super accessible to developers without advanced expertise. There were also several announcements regarding cloud video distribution through AWS Elemental, a service that will be incredibly helpful as more and more content creators and advertisers turn to video.
This was the first “real” day of re:Invent, and marked even more announcements. One of the most notable was AppSync, a managed GraphQL service that will be huge for mobile developers. Tuesday was also a big day for breakout sessions, and our own Craig Arcuri covered his wild journey to hear about five wildly different topics.
The first keynote was held on Tuesday night and featured Peter DeSantis, the VP of Global Infrastructure. The talk’s theme centered around “innovation at scale,” an apt description of AWS if there ever was one. We covered the keynote in detail here, including its two big announcements. The first was EC2 Bare Metal, a long-awaited option for running compute instances. The second was AWS GuardDuty, a new security service that uses machine learning to automate many of the toughest security tasks that organizations face. The conference’s first keynote combined exciting announcements with a (in my opinion) really cool look at the history of Amazon’s infrastructure. But it was just an appetizer for what was to come…
If re:Invent 2017 started with a bang, Andy Jassy’s keynote on Wednesday morning was an explosion. The CEO of Amazon Web Services announced more than 20 new services and features, and naming them all here just isn’t possible. Luckily, we’ve already done it – check out our full recap for details. To give a very brief summary, AWS announced services and improvements in five key areas:
- Big Data
- Machine Learning
- IoT (Internet of things)
Aside from databases, which have been a staple of “traditional” architecture for many years, these categories (note: these represent how I chose to sort the announcements) are indicative of something significant. Amazon is betting big on the role of the cloud in the way we think about technology. The popularity of AWS as a platform probably renders this idea self-evident, but I’ll quote Andy Jassy to illustrate what I mean: “We are going through the biggest transformation of technology in our lifetimes with the transformation to the cloud.” The power of the services AWS announced on Wednesday is proof of this. Not only is Amazon betting on the cloud – they’re a driving force behind its power. And with AWS, we no longer have to wait for the future to start leveraging this power to change the world. We can start today.
On Thursday, we heard from Werner Vogels, the CTO of Amazon Web Services. The theme of the day was “21st Century Architectures,” another choice that made a whole lot of sense in the larger context of the conference. The main takeaways from Werner’s keynote included:
- Voice service is the next big disruption
- Security is everyone’s job
- Reliability is a business choice
- The future is in the way AWS has built the cloud
Each of these talking points made for a great discussion (you can read a more in-depth analysis in our original post), but the last one was the most important in the greater context of re:Invent 2017. Amazon, again, has made it very clear that they are not content to just sit by and improve the way we use the cloud now. Their goal is to fundamentally change it.
Several new services were announced on Thursday as well, which we covered in our full recap of the keynote. A couple are worth mentioning again, though – specifically, AWS Cloud9 and the Serverless Application Repo. We’ll talk about what these mean in a moment, but they further support a fundamental shift in the way cloud computing affects the way our organizations run and the way we build things.
Analysis of re:Invent 2017
There’s a lot to say about all of the announcements made at this year’s conference, but a few key points emerged. Let’s talk about what the future of AWS holds, based on what we saw last week.
Amazon doubles down on serverless
Serverless has been a hot buzzword in cloud computing for a while, but it’s also much more. It makes computation easier, faster, and cheaper. When compute power is cheaper, it becomes more accessible, making the development of a broader range of applications and services much closer to reality.
Our resident serverless expert, Fernando Medina Corey, gave some excellent insight into this trend in our team’s Wednesday recap. Across all service categories, AWS showed a trend of managed services that they’re calling serverless because of the ways they work on the backend. Cost is a big factor, but what’s even bigger is the implication on the development and user innovation. Managed services are ready to be implemented with minimal setup, and often all you need to do is plug them into existing applications.
To give an example, Andy Jassy broke down machine learning into three layers: frameworks and interfaces, platform services, and application services. Platform services like SageMaker have implications at the organizational level and let’s be honest, SageMaker looks incredibly powerful. But AWS also released a number of application services as well. Services like Transcribe, Translate and Comprehend need little more than an API call to start using right away in existing applications. The power of these application layer services really can’t be overstated: individual developers can start using these services right away.
Focus on building
Amazon is putting a huge focus on “building.” There’s a lot of overlap between this concept and their focus on serverless (managed) infrastructure, but I think this is worth talking about separately.
In the final keynote, Werner Vogels touched on this quite a bit. AWS is an infrastructure provider, sure, but their choices of how they’re offering and implementing new services abstracts away much of the management. Think about what businesses do – they offer a service of some kind. That service runs on infrastructure, but its effectiveness from a customer perspective relies primarily on the service itself. By abstracting infrastructure through managed services, organizations are able to focus on building the things that actually generate revenue and make a difference in the lives of their customers.
It’s not all about business, though. Services like Cloud9 make development easier for individuals as well. As the number of self-taught developers continues to grow, access to enterprise-level tools on a personal scale becomes more and more important. From an organizational perspective, this increases the talent pool. But for the people who are using this technology, there are even greater benefits. Access to high-quality tools improves people’s ability to change their lives with technology. Greater accessibility means greater diversity in the tech industry. When people have the ability to use technology to solve the problems they face in their day to day lives, we all win, and that’s one of our core beliefs at Linux Academy.
The machine learning revolution
We’re in a golden age of machine learning, and Amazon, as usual, is innovating in what it believes will be the future of cloud computing. Services like SageMaker aim to bring machine learning to organizations everywhere, which just hasn’t been possible at a large scale in the past. Five or ten years ago, businesses that used machine learning and artificial intelligence were considered technological leaders. Increasingly, however, these technologies have become a requirement – as decisions become more and more data-driven, we need better ways to analyze and make sense of the vast amounts of data we now have access to.
There’s also a competitive element to this trend. Google and several other companies have arguably had an edge on Amazon in machine learning and artificial intelligence. The self-driving car division of Google, Waymo, is one example that’s cited often, and Google search has been head and shoulders above the competition for decades now. A few news outlets have speculated that the heavy focus on machine learning at re:Invent indicates a venture into these territories for Amazon. It’s hard to disagree, given the announcement of services like Comprehend and Translate, and time will tell just how well this strategy pays off.
Now that AWS re:Invent 2017 is in the books, we’ve got a lot to look forward to. We’ve covered most of them, but you can find a full list of services announced at the conference here. With so many new services, we can’t help but think about how much new information there is to learn. If you’re as excited about this as we are, we invite you to sign up and start training with us today – we’d love to help you on your cloud journey.
We’d also like to thank everyone who stopped by our booth to chat. Student success is our number one goal, and we love to hear from those who have been able to change their lives through cloud training.
If you were one of the lucky attendees, we hope you enjoyed re:Invent as much as we did (and if you weren’t able to make it, we hope you’re just as excited about all the new services). Either way, our team had an amazing time at last week’s event, and we hope to see you in Las Vegas next year!