There is a saying that every single year a couple of small businesses or growing startups are going to die around the same time. The carnage happens close to the end of the November each year. Amazon Web Services organize an event called re:Invent, where they announce significant changes in their ecosystem and platform - including new services and extensions for existing ones.
Jokes aside, keeping up with the pace of the changes across the year is incredibly hard. Over the year (including the conference days) Amazon Web Services announces several hundred big and small updates. We are doing our best to be up to date, as a part our professional interests, but also the passion.
Our team members - including our AWS Certified Experts - have chosen a subjective subset of announcements, that are the most interesting from re:Invent 2018.
We focused not only on the technical details and raw list of announcements, but they have included justification why that particular update brings value and new opportunities to the table.
Grab a cup of coffee (or favorite beverage), as we would like to present you our first announcement.
We recommend you a … recommendations engine
Finally Amazon starts to share their massive capabilities that are powering internals one of the biggest online retailers in the world.
One of such is Amazon Personalize which allows you to design, build, scale, and maintain recommendation and personalization engines. Everything almost effortlessly, with elasticity in mind, and no CAPEX and low OPEX.
Such a trend is visible during this year’s conference because that was not the only service built based on their in-house experience.
You can read more about the inner details of this service here:
When an evolution brings a revolution
Amazon SageMaker Neo is another step in the evolution of Machine Learning capabilities of the AWS platform.
Thanks to the primary service (Amazon SageMaker) we do not have to worry about building infrastructure, scaling and we can lower down OPEX metric when it comes to the maintenance. It takes care of optimization and deployment as well.
Neo is a step further - it provides model optimization for a given runtime. It not only shows the logical evolution in supporting other platforms like NVIDIA or ARM, but it gives a competitive advantage - because more platforms, means more opportunities to leverage ML concepts. That is a wise move, taking into account how much data we are producing at the moment in the placed called edge computing - including autonomous vehicles or anything related with internet of things.
Detailed announcement is here.
Developers, developers, developers…
Thankfully, there was no Steve Ballmer at the stage this year, but CTO of the Amazon Web Services - Werner Vogels - do not forget about developers.
Finally, we have a dedicated and robust set of tools for developing our lambdas locally, that can leverage our favorite IDE. I have chosen IntelliJ, but they announced similar plugins for Visual Studio Code and other JetBrains tools - including the one above.
Outside of typical development, it integrates AWS toolchain and allows to invoke and debug locally AWS Lambda implementation, with bells and whistles (no additional elements required).
Keep in mind that it is in developer preview - you can find more about it here.
As functional programmers, we are delighted that AWS Lambda runtimes are open from now on.
We were so excited that we tweeted that immediately:
Do you remember @afronski #CTA after his talk #FunctionalProgramming in #Serverless world? The one about improving the state of tools and support - looks like the #AWS listens everywhere. 👂 #Lambda #elixirlang #erlang #reInvent2018 #reInvent https://t.co/bmZhwEgBry— Pattern Match (@patternmatchPL) November 29, 2018
Now we can bring our favorite choices into the AWS Lambda without any additional hassle. Finally, we can write a correct Function as a Service solutions leveraging functional thinking as an implementation workhorse.
More about custom runtimes (and also layers which can help to reuse your code) you can find here.
A pragmatic touch for S3
Amazon S3 Intelligent Tiering is a huge thing. Knowing how much data our clients are storing in that service every solution that allows for cost reduction, better automation and enforcing conventions is more than welcome.
In the significant storage deployments, including data lakes, amount of data and users is exceptionally high. Designing and enforcing lifecycle policies in such environments it is not an easy task. Thanks to the new announcement it should be much easier from now on.
I want to emphasize how mature AWS approach is here. As a provider, they are cutting off their revenue by introducing such service, but that builds trust between provider and users. Also, it allows reinvesting saved money into new initiatives. Brilliant move! They left greed to the others.
Detailed blog post about how it works is available here.
Client-driven Blockchain Services
Here in Pattern Match we firmly believe that properly applied blockchain technology brings benefits and introduces revolutionary options.
Last year many people criticized AWS for not joining the blockchain bandwagon. They did not announce any new service related to that. This year they have made clear why.
Amazon is proud from its client obsession and dedication - and that also influences AWS. After listening to their customers and investigating their real needs, they have introduced two new, client-driven services.
For those that are seriously considering leveraging power of Hyperledger or Ethereum networks we want to introduce the new service called Amazon Managed Blockchain.
It provides two flavors mentioned above, including scalability, administration tools, certificates, metrics, and monitoring.
If you are interested in how it works, you can read about it here:
Announcements are overwhelming, but it challenges us!
Last, but not least - a warning and constructive opinion from Radek.
It is straightforward to be overwhelmed by that amount of news. It is also impossible to be always up to date - what we can recommend is to focus on your area of interest and stick to it. Listen very closely about updates for those services and slowly build your knowledge around the topic.
So that a year after, you will be ready for another avalanche of the updates. Until then, choose your niche and learn slowly about stuff that interests you.
Feel free to share your favorite announcements. For sure we missed a one or two, so do not hesitate to drop your favorite ones in the comments.
Focus On Product And Move Tech Out Of Sight
Amazon Web Services is our core expertise. Partner with our experienced and pragmatic builders that help you innovate and build a competitive advantage to move quicker and be cost-effective with use of AWS cloud platform.Schedule a call with our expert