everythingpossible - Fotolia
Did you miss the chance to attend AWS re:Invent? Or, like many others, could you not go because it was sold out?
I’ve noticed that most of those who deploy AWS solutions don’t have time to go to conferences. These days, most provider conference attendees are still at the research phase of their cloud journey. AWS did a good job of getting the word out at the conference about its new services, and AWS continues to prove its dominance in the still emerging IaaS public cloud computing space.
However, the world of public cloud computing has changed a great deal since the last re:Invent conference. Google and Microsoft got their respective cloud computing acts together, and both are coming on strong in the IaaS space. What’s more, enterprises now look at hybrid and multi-cloud as their best path to leveraging cloud-based platforms. On the downside, the emerging security and privacy issues continue to loom large, and news around the NSA scandal shows no signs of falling out of the news cycles.
That being said, AWS has not taken its foot off the accelerator. The news from re:Invent was powerful, as expected. Here are the three most important things you should know from that conference:
First, AWS announced the launch of Amazon Aurora cloud-based database service. This is a database built for cloud-native applications, with foundations in MySQL. Aurora is a managed database that exists under RDS, and, according to AWS, can support 6 million inserts per second and 30 million selects per minute. The storage is there as well, with 5 times more storage and twice the RAM disk of RDS, as well as very tight integration with S3. You’ll pay 29 cents per hour to use this service, and it’s not in the “preview” phase of the launch.
So, why is this a big deal? Aurora may replace the current MySQL engine (RDS MySQL) that exists now within RDS. Thus, it won’t make sense to leverage the old service, if the new one is so much better.
Second, AWS pushes further into the container world with Amazon EC2 Container Service (ECS). This is a hosted container management service that runs on top of Amazon EC2. This service can centrally manage running containers, including scheduling, and you’re able to access this service using a well-defined API. This also allows you to integrate with other AWS features, including EBS volumes and security groups.
While many felt AWS came late to the Docker party, the reality is that AWS customers can already leverage Docker through Beanstalk, as well as AWS instances. However, Google clearly beat AWS to the managed container game with Kubernetes, which is the path chosen by most major cloud providers. By not following the pack, AWS could provide a Docker option that they control exclusively, or could be left behind as the market chases Kubernetes.
Finally, AWS will support distributed computing with event management. AWS announced an event management system called AWS Lambda that will allow developers to capture and process events by triggering scripts, without the need to spin up VMs to host the code. This requires no additional servers or compute nodes to provision or configure for the service to function.
While this seems too complex for the non-developers, those charged with building distributed applications on AWS, even using Docker containers, will find that these services make a great deal of sense. The ability to manage events across AWS leads to a huge number of potential use cases that will make the AWS platform that much more valuable.
What’s clear from re:Invent is that AWS continues to be the public cloud to beat. While many predicted the cloud giant would stub their toe at some point, AWS has not done much wrong in the last several years. They understand their market, and have moved quickly to provide the cloud services to make their customers successful. I suspect it will be the same story at the next re:Invent, but a year is a very long time in the cloud computing space, and a lot can change.