By Aaron Cox
As you settle in to read this article I consider it an accomplishment, I have likely written the latest and last recap to 2018 re:Invent that someone would dare publish, it might be a record. In fact I challenge anyone to find one published later, if you do, I may or may not give you the crumpled dollar that may or may not be in my left pocket. The timing of this article is actually purposeful, re:Invent is like a firehose, but a firehouse the size of a skyscraper. I wanted some time to consider and reflect on the conference and provide input that was more than a regurgitation of the list of events. In addition its timing between the holidays and year end makes recap articles a lower priority for those of us that don't write for our day job. In fact, AWS if your reading (please be reading) great conference, bad time of year.
Let me start with, re:Invent is a great conference, I have been to several conferences over the years and none of them have shown the efficiency, value, and opportunity available at re:Invent. If AWS is central in your future, your attendance will be valuable, and if you can't attend you can access almost all of the sessions online afterwards. Make the time, you will learn something valuable.
From the moment I landed in Las Vegas the efficiency was center stage, the registration line was immense, there were nearly 100 people ahead of me and I was in line for far less than 10 minutes. This is a fair example of every line I was in at the conference, pretty impressive for such a crowded event. Although I am not a conference swag hound I offer a high five to those who do the swag for the conference. Quality gear I actually use, I think that is a first. With that being said, be certified, the lounge is worth it and you get a bag full decent gear to boot. Actually don't be certified, I don't need a bigger crowd in the lounge eating my snacks and stealing my soft chairs.
The landscape of the conference is huge, the sheer number of venues is tough to tackle, my suggestion to you is try to stay at the same venue the entire day. I lost a lot of time moving between venues, this was a lesson learned for me. I was warned more than once not to spread my sessions and customer meetings too far apart, I made it work, but won't be doing that again.
Go for the labs and attend sessions when you need a break. I am not kidding and I am not saying that because the seats in the best sessions all book up 24 hours after the schedule is released. If you are there to learn and improve your technical skillset the various labs, trainers, and the access to them is extremely valuable. If you are serious about learning at the conference invest the time, the labs available range across all skill sets. They are also a fantastic networking opportunity, you have to try to not meet people when you spend time there.
The keynotes were great, watching them in the overflow was even better, unless of course you like being herded like cattle to a seat not of your choosing and enjoy sitting wedged between software developers that are too big for the small chairs they were seated on. If that's your thing then the main hall is for you. However the content was worth it even if that described your seating scenario. Andy Jassy does not pull punches. I have not seen that level of trolling from a keynote session EVER, Andy and Werner are coming for your traditional technologies, they will not be taking prisoners or concern themselves with traditional social cues. They intend to invest heavily to stay ahead of the pack in innovation. If you have not watched the keynote, even if you hate cloud, AWS, and everything it stands for.....it will still be worth the time strictly on entertainment value. I am not saying other providers won't keep up or are not innovating, however AWS is taking innovation personally and have no concern who or what they disrupt in the process. Outside of Google who is still catching up on scale of services (they are doing fine on their own innovation path) the other providers have businesses and footprints that must be protected until the balance of their cloud investment outweighs their legacy investments and Microsoft is barreling down that path as fast as they can.
As always the stream of product releases was impressive, there were some I saw as immediately impactful and others as positioning to get ahead of other providers. There are some I am qualified to discuss and we'll touch on those, while there are others I will likely never be qualified for like AWS Ground Station, their entry into fully managed satellite services, and yes I am talking about the kind of satellites orbiting the earth. Not sure I even thought of that as a market looking for a managed service to buy. I have a feeling there are a handful of companies that I have never heard of very unhappy about Amazon getting into this space.
Lets start with their Managed Blockchain announcement. The industry has generally accepted the potential and the value of Blockchain, to the point it is now a buzzword that many don't understand. Several organizations have struggled to deploy functional and reliable Blockchain capabilities. Having access to a reliable service that enables organization to ease support and deployment of blockchain is likely the helping hand Blockchain needs to become more widely used and successfully deployed.
For the average organization Robomaker and Deep Racer may not seem like valuable releases. However the impact of these releases will be felt over time. Robomaker provides easier access to smaller organizations that want to dip their toes into robotic software development. A whole industry is quickly coming to fruition in this space and access to these tools will speed the growth of this industry and provide broader access to entry.
Machine Learning is rapidly finding its way into our industry, but the speed of adoption will only move as quickly as the required skills are available. Although there were several exciting services around machine learning at re:Invent Deep Racer may be the most important, what Amazon is building around this product is going to offer an exciting and fun avenue to skill development and increase the available talent pool quickly. Deep Racer won't directly change how your organization does machine learning, but it will change the talent pool and therefore adoption. This is a great opportunity to dive into machine learning if that's a direction you want to take your career.
AWS Outposts was a popular discussion item the minute it was announced. AWS wasn't just happy with their attempt to host all of your workloads, now they seem to want to sit on your data center floor too. I am less excited about this release than others, yes it gives an onramp to their services for slower adopters and easier access to key cloud services in your data center, but I have not seen the level of impact from Azure Stack that was expected and would require Amazon to get into this space. It appears to be a functional and capable service that use cases exist for, however I am unsure we will see adoption rates that make this an impactful capability for them.
AWS Transit Gateway may be one of the more impactful releases for large enterprise deployments and for customers who have used a highly segregated VPC architecture. VPC peering in complex designs has been a bain of the existence of many cloud architects and network engineers. Although some were hoping for an announcement with even greater transitive peering capabilities Transit Gateway will now allow for you to centrally manage peering across VPCs and accounts providing a single hub to attach VPC and onsite connected networks. This is extremely impactful to easing the pain of design and support of complex designs. This may prove to be one of the most impactful yet least talked about releases.
I probably say it at least once a week if not more, you cannot handle configuration management in the cloud the same way you handled it in the data center, the sins of this function being a secondary priority will absolutely ruin any value in the cloud. Amazon has worked to provide tools that help with this transition however the release of Control Tower is a huge step towards using the power of infrastructure as code to baseline key configuration management details at the moment of deployment that can then be enforced through policy long term. Additional releases like Cloud Map and App Mesh also ease the effort in what is often a new priority for cloud adopters.
Security in the cloud is always a key discussion for new cloud adopters and vigorous cloud avoiders. Amazon continues to release quality tools and services in this space, however they have traditionally run independent of each other. Security Hub now brings an organizational view to monitoring and securing your environment integrating existing services through this single service. I find one of the most difficult aspects of securing cloud has been the difficulty in aggregating all the information and capabilities available to you. I need more time to see this service in action to determine if it will be as impactful as I hope, however this is a step in the right direction to gaining better control for your security requirements.
Lastly lets talk about the Partner Summit. I work for an AWS Consulting Partner and the partner program is important to us and other firms to help demonstrate our skill and capability in AWS to our clients just as it is for other partners. At re:Invent Amazon changed the partnership qualification drastically, most boutique firms I spoke with after the announced changes were disappointed and concerned about the changes, even some larger firms were unhappy. Personally I see it as a good thing. Previously a firm could have one or two people who would amass certifications and experience however there was no focus in building the expertise of the entire firm. This was typically the method used to climb through the partner levels for firms of all sizes. This meant that customers would engage a partner and often never work with certified resources. Amazon wants to ensure that partners can provide validated and capable resources to their clients, now the number of certified individuals is counted rather than the number of certifications. For firms that recognize skill development and experience are critical to successful cloud consulting and customer adoption, this was a positive. At CloudLogic we saw this as a positive as we require all of our people to be certified in more than one cloud provider and all consultants carry at least one AWS certification.
So what does this mean for you the individual, AWS is placing more and more importance on certification, which means consultancies, managed service providers, and even your traditional IT organization will increase their reliance on certification as a tool to determine qualified employees. If you are serious about your cloud skillset now is the time to certify if you have not started already. In addition as the Partner program sorts itself it allows customer to have more confidence on the of AWS partners.
So is re:Invent worth the time and expense? Absolutely! I have already marked off my calendar for next year. The value you gain from the conference will be heavily measured in how seriously you take the time to plan your time and take advantage of what you need from the conference. This is an event that your investment into it will increase the return you receive from it.