Attending technology events is always a targeted objective for me to know the pace of technology around the world. AWS re:Invent/Recap is one of those event which are always pinned on my priority board and it's not because it's one of the top tech event but for their advanced, automated, intelligent and, some times, crazy announcements every year. Recently, I have attended a re:Invent Recap event and gathered a lot of information on new AWS services and features. In this post, I will highlight some of those key announcements done by AWS in December 2019 and for full review visit AWS re:Invent.

First of all, for the audience who doesn't know about Amazon Web Services (AWS) or AWS re:Invent event:

  • AWS is a first public cloud service provider to launch the service in 2004.
  • AWS is world's largest cloud provider with 47.8% market share in IaaS (Infrastructure as a service). (Source: Forbes-Gartner)
  • AWS re:Invent is one of the world's largest tech event which started in 2012 with couple of thousand attendees v/s 65,000 in December 2019.
  • The event in December 2019 had delivered 3,898 technology sessions in 5 days.

Let's get back to the agenda, here are some key takeaways from the event:

Compute: 

  • AWS Graviton2: ARM based processor to deliver best prices performance of EC2 instances and a successor to AWS Graviton first generation.
    • Benchmarked to deliver 7x performance, 4x compute cores and 5x faster memory when compared to the first-generation Graviton processors.
    • Powers Amazon EC2 M6g (General Purpose), C6g (Compute), and R6g (Memory) instances that provide up to 40% better price performance compared to current generation instances.
  • AWS Nitro Enclaves: Hypervisor layer enabling customers to create isolated compute environments to further protect and securely process highly sensitive data within their Amazon EC2 instances.
    • Some of the Apple users may know about enclaves in their devices to protect sensitive data. If not, you can read it HERE.
  • Amazon Braket: Do want to solve Quantum mechanics computing issues? Try Amazon Braket for fully managed Quantum computing services.
  • AWS Compute Optimizer: ML-Powered instance and auto-scaling recommendation engine to recommend on how to optimize your instances.
  • AWS Fargate: Run serverless Kubernetes containers securely, reliably and at scale on AWS.

Storage:

  • Amazon S3 Replication Time Control: Provides a predictable replication time backed by a Service Level Agreement (SLA).
  • Amazon EBS Faster Snapshot Restore (FSR): Keeps the snapshots warm for faster EBS volume creation.
  • Amazon S3 Access Point: Stop sharing single ARN for S3, create hundred of access points per bucket and simplify managing data access at scale for applications using shared data sets on S3.

Databases:

  • Amazon Managed Apache Cassandra Service (MCS): Amazon know how to make you lazy (a.k.a. focus on application)  with their managed services and now, it's Cassandra. Cassandra not being a popular choice in many parts of Europe but highly demanding in US and also bags, one of the biggest IT use case of Netflix.
  • Amazon RDS Proxy: Fully managed, highly available database proxy feature for Amazon RDS.

Analytics:

  • AWS Data Exchange: Easily find and subscribe to 3rd-party data in the cloud.
  • AQUA for Amazon RedShift: AQUA (Advanced Query Accelerator) is a new distributed and hardware-accelerated processing layer to run queries 10x faster than any other cloud data warehouse.

Management Tools:

  • AWS CloudTrail Insights: Identify unusual activities while providing advanced insights such as spikes in resource provisioning, bursts of AWS IAM actions or gaps in periodic maintenance activity.

Security:

  • Amazon Detective: To analyze, investigate and identify the root case of security findings and suspicious activities.
  • AWS IAM Access Analyzer: Continuously ensure that policies provide the intended public and cross-account access to resources such as Amazon S3 buckets, AWS KMS Keys and AWS identity and Access Management tools.
  • AWS Single Sign-on: Create or use existing identities, including Azure AD and manage access centrally to multiple AWS accounts and business applications.

Networking:

  • AWS Transit Gateway: New features of Transit Gateway has brought a big relief for network team to help them create a unified network. With these new features, AWS has enabled building efficient global networks:
    • Inter-region Peering: Build global networks connecting multi AWS regions
    • Multicast: Build and deploy multicast applications in the cloud
  • Amazon VPC Ingress Routing - Route inbound and outbound traffic through a third party or AWS single appliance.

Serverless:

  • Provisioned concurrency on AWS Lambda: Let your lambda functions ready to run code with warm containers. This will keep functions initialized and hyper-ready for execution.

Infrastructure:

  • AWS Outposts: Fully managed service that extends AWS infrastructure, AWS services, APIs and tools to virtually any connected customer site. A consistent hybrid experience for applications across on-premise and cloud environment. AWS Outposts supports AWS EC2, EKS, App Mesh, EMR, RDS and S3 (not yet GA).
  • AWS Local Zones: Extend the AWS cloud to more locations and closer to end-users to support ultra low latency. First AWS Local Zone will be located in Los Angeles.
  • AWS Wavelength: Embeds AWS compute and storage inside telcos providers' 5G networks.

Today, none of the tech event is complete without Artificial Intelligence and Machine Learning announcement. So, let's see what AWS has new for their customers/users in the field of intelligence.

  • Amazon Sagemaker: There was a lot of features added for Amazon Sagemaker as follows:
    • Amazon Sagemaker for Kubernetes: Customers with Kubernetes can now train, tune and deploy models in Amazon SageMaker.
    • Amazon Sagemaker Studio: Fully integrated development environment (IDE) for machine learning.
    • Amazon Sagemaker Notebooks: Fast-start shareable notebooks to ease collaboration and flexibility for engineers and data scientist.
    • Amazon Sagemaker Processing: Analytics jobs for data processing and model evaluation
    • Amazon Sagemaker Experiment: To organize, track and evaluate training experiments.
    • Amazon Sagemaker Debugger: Analysis, debugging, explainability and alert generation on models.
    • Amazon Sagemaker Model Monitor: Continuous monitoring of models in production.
    • Amazon Sagemaker Autopilot: Automatic model creation with full visibility and control.
  • AWS DeepRacer Improvements: AWS DeepRacer was announced in 2018 where developers gets an opportunity to get hand-ons with a fully autonomous 1/18th scale race car driven  by reinforcement learning, a 3D racing simulator and a global racing league. This event had announced improvements on AWS DeepRacer where developers can create their own races, have object detection/avoidance and do head-to-head races
  • AWS DeepComposer: I have talked about this services, in one of my previous post where AWS had announced world's first machine learning enabled musical keyboard. So now, if you are a developer and want to be a good musician then AWS DeepComposer is for you. Who knows, if you can become a disciple of great "Ludwig van Beethoven".
  • Amazon Fraud Detector: A fully managed service that makes it easy to identify potentially fraudulent online activities such as online payment fraud and the creation of fake accounts.
  • Amazon CodeGuru (in Preview): A machine learning services for automated code reviews and application performance recommendations. It also helps you identify your most expensive lines of code.
  • AWS Inferentia: Last but not the least, high performance machine learning inference chip custom designed by AWS to support upto 128 TOPS (trillions of operations per second).

Note: Most the given services or features in this post are cost driven and therefore, it is recommended to read the documentation before you start playing around it.

At the end, I can say that AWS announcements for 2020 were driven towards automation and latency improvement with better price performance.