Amazon’s flagship annual event for the global cloud computing community AWS re:Invent 2020 has been virtually conducted this year. There are several keynote sessions, product announcements, product deep-dives, free up skill training and certification, cloud computing best practices, and leadership sessions for more insights into all AWS offerings for two weeks.
Summarized in this blog are my top picks from the keynote session by Andy Jassy, CEO of Amazon Web Services (AWS). Andy has addressed the latest trends and challenges for the businesses and announced new advancements, technologies, and AWS products.
Need to reinvent
“Only small percentages (10.4%) of Fortune 500 companies in the past remain on that list today”. Businesses that survive and thrive continuously reinvent themselves to adapt to the market disruption, satiate customer demands, and win over the competition. The keys to reinventing include:
- Leadership’s willingness and courage to acknowledge and dig into business data, analyze performance, and take new steps in the right direction.
- Leaders need to have an honest view of the marketplace – competitors, their products/services and make changes desirable.
- Leadership needs to continually hire new talent who can provide fresh perspectives, which old-timers often can’t.
- Listening to the customer needs to be of paramount importance. The market feedback should act as the guiding force behind all business reinvention.
- Businesses can’t wait to reinvent, and they should act fast and not be afraid to fail. The failing process should be quick so that they learn from their mistakes and keep improving.
- Business transformation is complex, and to keep things simple for embracing your technology needs, choose a primary partner to gain momentum, view results, and take the next steps.
- Businesses should inculcate a sense of reinvention culture and use integrated platforms. AWS has been a consistent market leader (revenue run rate of $46 billion in 2020.) by sharing a broad range of AWS services and solutions to meet customer needs.
- Having understood the need to reinvent, businesses must drive forward at speed with aggressive top-down goals. By setting aggressive targets initially, General Electric successfully migrated thousands of workloads to AWS in just a couple of years.
Further, Andy Jassy’s keynote included several announcements for new partner offerings, new product lines, and extensions, some of which will be generally available only in 2021. Among the vast list of announcements, some are mentioned below:
New Products, Extensions, and Services
- Amazon EKS Anywhere (upcoming): Deploy Kubernetes clusters in your datacenter. Read more.
- Amazon ECS Anywhere (upcoming): Deploy native Amazon ECS tasks in the traditional AWS managed infrastructure and in your datacenter as well. Read more.
- AWS Proton: Overcome the challenge of maintaining microservice with changing infrastructure and resource configuration with AWS proton. Superior aid to automate and manage infra provisioning. Read more.
- AWS Lambda Container Support: Easy way to provision Lambda functions maximum of 10,240 MB (10 GB) of memory. This is a help for batch processing, ETL jobs, media processing, etc. Read more.
- Amazon Elastic Container Registry (ECR): Easy and highly available process to publicly share and deploy container software. You can learn more about Amazon ECR here.
- Amazon EKS Distro: It is an open-source Kubernetes distribution used by Amazon EKS to help create reliable and secure clusters. Read about Amazon EKS Distro
Database and data management
- AWS Glue Elastic Views (preview): Combine and replicate data across multiple data stores without writing custom code using AWS Glue Elastic Views.
- Babelfish for Amazon Aurora (preview): Migrating from on-premises facilities to the AWS cloud means moving from Microsoft SQL Server to PostgreSQL. A considerable effort is required to update applications and queries. Babelfish for Amazon Aurora is a new translation layer that enables Aurora to understand questions from applications written for Microsoft SQL Server.
- Aurora Serverless v2 (preview): Based on your application’s needs, Aurora Serverless auto-scaling configuration for Amazon Aurora can automatically start up, shut down, and scale capacity up or down. Aurora Serverless v2 can provide almost 90% cost savings compared to provisioning for peak capacity. Ain’t that cool!
- Amazon EBS gp3: EBS gp3 are next-generation general purpose SSD volumes for Amazon Elastic Block Store (Amazon EBS) that enable you to provision performance independent of storage capacity and at a price point that’s as much as 20% lower than existing gp2 volumes.
- io2 Block Express EBS Volumes: Larger and faster io2 Block Express EBS volumes with higher throughput and maximum volume size of 64 TiB, and above all with sub-millisecond, low-variance I/O latency.
- Tiered pricing for IOPS: Amazon announced tiered pricing for input/output operations per second (IOPS) charges for Amazon Elastic Block Store (EBS) io2 volume, reducing the cost of provisioning peak IOPS by 15%.
- Amazon DevOps Guru: Amazon DevOps Guru is an end-to-end managed operations service that automatically collects and analyzes data to identify behavior regressions and suggest quick fixes. DevOps Guru automatically fetch data from sources such as Amazon CloudWatch, AWS Config, AWS CloudTrail, and AWS X-Ray, and used machine learning to analyze the disparate data. It monitors the metrics to establish patterns in system behaviour and performance, and notifies if there are deviation from normal operating pattern so that you can fix it before it can impact your customers. Read More.
- AWS Trainium (available in 2021): AWS Trainium is a custom made, high-performance machine learning training chip available as an EC2 instance inside Amazon SageMaker.
- Habana Gaudi-based EC2 Instances (available in 2021): The new Habana Gaudi-based EC2 instances will leverage up to 8 Gaudi accelerators and deliver up to 40% better price-performance than current GPU-based EC2 instances for training deep learning models.
- New Amazon SageMaker capabilities: SageMaker Feature Store and SageMaker Data Wrangler will reduce the time it takes to aggregate and prepare data for machine learning from weeks to minutes.
- Amazon QuickSight Q (preview): QuickSight Q is a Natural Language Query (NLQ) feature powered by machine learning that enables business users to ask questions about their data using everyday language and receive accurate answers in seconds.
- Amazon Connect Wisdom: Amazon Connect Wisdom uses machine learning to drastically reduce the amount of time that call center agents spend searching for answers.
- Amazon Monitron: Amazon Monitron is an end-to-end system that detects abnormal behavior in industrial machinery, enabling you to implement predictive maintenance and reduce unplanned downtime.
- Multi-destination S3 Replication: elastic, fully managed features so you can automatically replicate objects between buckets. Learn more here.
- Updates to Amazon S3 Intelligent-Tiering: Amazon S3 Intelligent-Tiering, a storage class that automatically moves data to the most cost-effective access tier without impacting performance, has been updated with archive and deep archive storage.
- Amazon S3 Bucket Keys: Reduce request costs for server-side encryption (SSE) with AWS Key Management Service (KMS) by up to 99% by decreasing the request traffic from S3 to KMS. Learn more about Amazon S3 Bucket Keys here.
- Amazon EC2 Mac Instances: Amazon Elastic Compute Cloud (EC2) Mac instances for macOS are generally available on November 30, 2020. Built on Apple Mac mini computers, EC2 Mac instances enable customers to run on-demand macOS workloads in the AWS cloud for the first time.
- New Amazon EC2 Instances: R5b, M5zn, D3/D3en available now, with T4g available for a free trial, C6gn, and G4ad instances coming soon.
This is a highlight of the key AWS product offerings that I thought is relevant to anyone savvy with cloud-native applications, but there are still many more exciting sessions where other new products and services announcements are highly probable.