The last AWS re:Invent 2019 was full of unleashes and announcements of new services, products and features of the services we already know. During five exciting days, full of innovation, AWS experts delved into crucial topics in this era of digital transformation such as machine learning, cloud computing, data storage and security.
At Linke, we collected the following AWS services highlights. Keep reading!
AWS Nitro System
This point was one of the most talked about in the AWS re: Invent. The AWS Nitro System is the underlying platform for our next generation of EC2 instances that enables AWS to innovate faster, further reduce cost for our customers, and deliver added benefits like increased security and new instance types. As a component of the Nitro system, the Nitro Hypervisor primarily provides CPU and memory isolation for EC2 instances.
AWS has the largest in-memory cloud instances for SAP Applications with 24TBs and 448 vCPUs that allow you to get an incredible performance up to 480k SAPs and a max storage bandwidth of 28 Gbps and max network of 100Gbps.
Amazon SageMaker Studio
Another important point during the AWS re:Invent were all the products related to Machine Learning. In this aspect, one of the products presented was the Amazon SageMaker Studio: the first fully integrated development environment (IDE) for machine learning.
Amazon SageMaker Studio unifies at last all the tools needed for machine learning development. Developers can write code, track experiments, visualize data, and perform debugging and monitoring all within a single, integrated visual interface, which significantly boosts developer productivity.
SageMaker studio includes the following tools:
- Amazon SageMaker Notebooks
- Amazon SageMaker Experiments
- Amazon SageMaker Processing
- Amazon SageMaker Debugger
- Amazon SageMaker Model Monitor
- Amazon SageMaker Autopilot
S3 Access Points
This is a feature related to the Amazon S3 product, S3 access point simplifies managing data access at scale for applications using shared data sets on S3. The advantages of this new feature are as follows following:
- Large shared data sets
- Restrict access to VPC
- Test new access policies
- Limit access to specific account IDs
- Provide a unique name
AWS made a series of talks during re: Invent about how to extract data from SAP and then be exploited by Data Lakes.
It was shown how to extract data from SAP using the following AWS services: Lambda, Glue and API Gateway. This data must be stored in the AWS S3 service so that they can then be exploited by AWS services such as Athena, Redshift or can be processed with machine learning.
6th Generation Families
The sixth generation of instances with the following characteristics was also presented:
- The new general purpose (M6g), compute optimized (C6g), and memory optimized (R6g) Amazon EC2 instances deliver up to 40% improved price/performance over current generation M5, C5, and R5.
- 4x more compute cores, 5x faster memory and 7x more performance with AWS Graviton processors.
- M6g is currently available and C6g and R6g are coming in 2020.
AWS Code Guru
Amazon Code Guru is an Amazon EC2 service for Machine Learning. It is used for automated code reviews and application performance recommendations. AWS Code Guru helps in the search of the lines of code that most damage the performance of the application and stays active to find a solution to the problem. Once the solution is found, it gives us specific recommendations to correct or improve the code.
The new presented features were:
- Inter Region Peering: Sends inter region traffic privately over AWS’s global network backbone improving the performance of your connectivity.
- Accelerated VPN: To route VPN traffic from remote locations through the closest AWS edge location to improve connection performance.
- Network Manager: To manage a global network across AWS and on-premises (Centralized Network Dashboard monitoring).
- Multicast: Building multicast applications in the cloud and distribute data across thousands of connected Virtual Private Cloud networks. It's the first Cloud with this functionality.
FSX for Windows File Server
AWS FSX enables easy and efficient migration of Windows-based applications that require shared storage on AWS instances. Highlights were:
- MultiAZ: Replicates data across Availability Zones and automatic failover providing a High Availability solution for different applications (SAP Business Objects and OpenText).
- HDD based storage: Lower prices
- Allows create SQL Server HA environments
And there's more...
As a curiosity, a world championship was held with cars on a scale of 1/8. These cars used the AWS DeepRacer service.
AWS DeepRacer gives you an interesting and fun way to get started with reinforcement learning (RL). RL is an advanced machine learning (ML) technique that takes a very different approach to training models other than machine learning methods.
We were in a workshop to learn how to use the AWS DeepRacer service. It was a very entertaining and curious workshop, since you could see the car live and you could train your model and once trained, go to another room to load your model in a car and be able to compete with your model in a real circuit.
One of the advantages provided by this service is that it is not necessary to own a 1/8 scale car to use this service, since it has a 3D simulator to train your models.