AWS Outposts (pictured above) puts AWS-managed compute and storage racks on customers' premises, making it the equivalent of Oracle's long-established Cloud@Customer offering.
AWS Outposts can be installed in virtually any datacenter, co-location space, or on-premises facility, and customers use the same AWS APIs, control plane, tools, and hardware on-premises as they do in the AWS cloud.
Customers can choose from a range of compute, storage, and graphics-optimised Amazon Elastic Compute Cloud (EC2) instances, with or without local storage options, and Amazon Elastic Block Store (EBS) volume options.
Other services that can be run locally include Amazon Elastic Container Service (ECS), Amazon Elastic Kubernetes Service (EKS), Amazon Relational Database Service (RDS), and Amazon Elastic MapReduce (EMR).
Amazon plans to add other services to that list, starting with S3 some time in 2020.
Private connections can be established to regional services including Amazon Simple Storage Service (S3) buckets and Amazon DynamoDB tables.
The AWS native version of AWS Outposts is available now, linked to the US East (N. Virginia), US West (Oregon), US East (Ohio), EU West (Ireland), Asia Pacific (Seoul), and Asia Pacific (Tokyo) regions. More regions are coming soon, according to the company.
VMware Cloud on AWS Outposts is expected to become available during 2020.
"When we started thinking about offering a truly consistent hybrid experience, what we heard is that customers really wanted it to be the same – the same APIs, the same control plane, the same tools, the same hardware, and the same functionality. It turns out this is hard to do, and that's the reason why existing options for on-premises solutions haven't gotten much traction today," said AWS vice president of compute services Matt Garman.
"With AWS Outposts, customers can enjoy a truly consistent cloud environment using the native AWS services or VMware Cloud on AWS to operate a single enterprise IT environment across their on-premises locations and the cloud."
AWS Fargate simplifies running Kubernetes applications on AWS by removing the need to manage servers and clusters.
AWS Fargate has been introduced because so many customers are running Kubernetes on AWS.
The Amazon EKS managed Kubernetes service made it easier to manage, scale, and upgrade Kubernetes clusters, but customers still need to patch servers, choose which Amazon EC2 instances to run on, patch the instances, scale cluster capacity, and manage multi-tenancy.
AWS Fargate for Amazon EKS combines the power and simplicity of serverless computing with the openness of Kubernetes. AWS Fargate automatically allocates the right amount of compute, thus right-sizing performance and cost.
It also provides strong security isolation for every pod by default, and eliminates the need to patch, scale and secure a cluster of Amazon EC2 instances.
"AWS Fargate has made it so much easier for Amazon ECS customers to manage containers at the task layer versus worrying about servers and clusters," said AWS vice president of containers Deepak Singh.
"Our Amazon EKS customers have been clamouring for us to find a way to make Fargate work with Kubernetes, and we're excited to do so today. With AWS Fargate, Kubernetes customers can truly take advantage of the elasticity and cost savings of the cloud when running their Kubernetes containers, and don't have worry about patching servers, scaling clusters, or managing multi-tenancy."
AWS Fargate for Amazon EKS is available today in US East (N. Virginia), US East (Ohio), Europe (Ireland), and Asia Pacific (Tokyo), with more regions coming soon.
National Australia Bank (NAB) executive general manager of infrastructure cloud and workplace Steve Day said "Amazon ECS has already reduced NAB's microservice development time by a factor of 10. With AWS Fargate for Amazon EKS, we expect to improve this even further by enabling low touch Kubernetes cluster management at scale.
"By removing the need for infrastructure management, we expect AWS Fargate for Amazon EKS to reduce our development costs on new projects by 75%. Over the next 12 months, migrating to AWS Fargate for Amazon EKS will enable 100 NAB service teams with a managed microservices based platform to break down 50 monolithic applications into modern architectures."
Amazon Managed Cassandra Service is a serverless, scalable, highly available, and fully managed database service that supports Cassandra workloads. It allows developers to use their current Cassandra application code, Apache 2.0 licensed drivers, and tools without having to worry about managing the underlying infrastructure.
Amazon Managed Cassandra Service is compatible with the open-source Apache Cassandra 3.11 API, and provides single-digit millisecond performance at any scale.
Tables can be scaled up and down automatically based on actual application traffic, and customers are only charged for the resources actually used.
In early 2020, AWS will provide customers with tools to migrate in-premises or EC2 Cassandra tables to Amazon Managed Cassandra Service.
"Customers who are running Cassandra have told us they want to spend less time scaling, patching, and managing infrastructure and more time building applications at scale," said AWS vice president of databases Shawn Bice.
"Amazon Managed Cassandra Service gives these customers the ability to run Cassandra without having to worry about managing the underlying hardware, and it gives them an easy way to integrate their Cassandra applications with other AWS services.
"Because Amazon Managed Cassandra Service is also serverless customers can stand up Cassandra clusters in minutes and scale their database up and down based on demand, so they can spend less time worrying about their clusters, and more time building the applications that impact their businesses."
Adobe principal architect Dan Neff said "We've used Cassandra for years to help us store and manage data for many of Adobe's applications. However, managing our Cassandra infrastructure is expensive and complex.
"With Amazon Managed Cassandra Service, we look forward to having a managed option to run our Cassandra workloads that offers us consistent performance at-scale and built-in enterprise management and security features, such as monitoring using Amazon CloudWatch, fully-managed durable storage, encryption, and access management using AWS Identity and Access Management (IAM)."
AWS Wavelength puts AWS compute and storage services at the edge of 5G networks, making it possible to build applications with single-digit millisecond latencies.
Without Wavelength, latencies between a mobile device and AWS resources can be more than 100ms.
Use cases for Wavelength are said to include machine learning inference at the edge, autonomous industrial equipment, smart cars and cities, IoT, game streaming, and augmented and virtual reality.
Collaborating carriers initially include Verizon in the US (a pilot is running in Chicago), Vodafone in Europe, SK Telecom in South Korea and KDDI in Japan, with availability in 2020. Other carriers are expected to join the scheme.
"With Wavelength, we bring 5G and cloud together to give our customers the powerful new capability to run cloud services consistently within a few milliseconds of mobile end-users," said AWS vice president of compute services Matt Garman.
"This is a game changer for developers that is going to unlock a whole new generation of applications and services. We are really excited to see our customers innovate with these unique new capabilities that they did not have access to before."
Amazon Redshift RA3 instances with managed storage are said to run three times faster than any other cloud data warehouse, and allow customers to separately scale compute and storage. These instances combine SSDs for local storage and S3 for durable storage, and high-bandwidth networking speeds transfer to and from S3.
AQUA (Advanced Query Accelerator) for Amazon Redshift (coming in mid-2020) is a new distributed and hardware-accelerated cache for Amazon Redshift, promising performance ten times faster than any other cloud data warehouse. AWS says this level of performance will be sufficient to allow raw data to be queried directly. AQUA-powered Amazon Redshift will be 100% compatible with the current version of Amazon Redshift, allowing migration with no code changes. .
Amazon Redshift Data Lake Export (available now) allows data exports directly from Amazon Redshift to Amazon S3 in the open Apache Parquet data format. This means the results of an Amazon Redshift query can be further analysed in services such as Amazon SageMaker, Amazon Athena, and Amazon EMR.
Amazon Redshift Federated Query (currently in preview) permits queries in Amazon Redshift on live data across Amazon Redshift data warehouses, Amazon S3 data lakes, and Amazon RDS and Amazon Aurora (PostgreSQL) operational databases.
Familiar SQL statements combine data across these various data stores.The Redshift query optimizer intelligently distributes as much work as possible to the underlying databases.
UltraWarm for Amazon Elasticsearch Service (in preview) simplifies the collection, analysis, and visualisation of machine-generated log data from websites, mobile devices, and sensors by providing a warm storage tier that is sufficiently cost-effective to store years of data while still delivering a snappy, interactive experience.
UltraWarm reduces costs by up to 90% to store the same amount of data in Elasticsearch today, and is 80% lower than the cost of warm-tier storage from other managed Elasticsearch offerings.
"Our customers tell us they are regularly dealing with petabytes, and even exabytes of data, and their existing analytics systems can't keep up," said AWS vice president of database services Raju Gulabani.
"These customers want to perform fast analytics on all of their raw data across their data warehouse and data lake, and cost effectively deal with the explosion in log data to retain information that might help them run their businesses better. With today's announcements we are helping AWS customers do all of this and fearlessly embrace data at scale."
Genealogy service Ancestry senior manager of engineering development Clint Smith said "With Amazon Elasticsearch Service we collect and analyse our company's operational logs in real time.
"Now UltraWarm for Amazon Elasticsearch Service will help us identify correlations between logging events and quickly root-cause application problems.
"Before UltraWarm for Amazon Elasticsearch Service, our cost constraints meant we could only store five days of data. With UltraWarm for Amazon Elasticsearch Service we will be able to extend that window to 90 days, and analyze the data via Kibana at a significantly lower cost. This extra data will help us identify application problems that we just couldn't see with the five days of data we were storing before."
AWS Contact Lens provides analytics for users of the Amazon Connect managed cloud contact centre service.
Coming in mid-2020, AWS Contact Lens gives contact centres the ability to understand the sentiment, trends, and compliance of customer conversations to improve customer experience and identify crucial customer feedback.
Customer service supervisors can discover emerging themes and trends from customer conversations, conduct fast, full-text search on call and chat transcripts to troubleshoot customer issues, and improve customer service agents' performance with call and chat-specific analytics – all from within the Amazon Connect console.
Contact Lens provides fully managed machine learning-powered analytics capabilities that do not require any coding or ML experience. Calls are automatically transcribed and indexed along with chat sessions, and analysed for content and sentiment.
Themes arising in multiple conversations are automatically surfaced.
"At Amazon, customer obsession drives everything we do. Over the years, we've developed unique expertise in using machine learning to better understand our own large volumes of customer contacts and take appropriate action," said AWS vice president of productivity applications Larry Augustin.
"Contact Lens brings together the technology and expertise Amazon has developed to support its call center operations and delivers it to Amazon Connect customers – without requiring any machine learning or programming expertise to use it. We are excited to see how our customers benefit from our experiences to improve their customers' trust and loyalty."
Publisher News Corp already uses Amazon Connect for its service desk. senior vice president of end user and infrastructure services Simon Clark said "We are now looking forward to using Contact Lens for Amazon Connect because its powerful features for both voice and chat interactions will make it possible for our contact center staff to provide a better experience for our employees by seamlessly leveraging the power of machine learning."
MVNO amaysim is another Amazon Connect customer. IT operations director Peter James said "Contact Lens for Amazon Connect not only provides us with an end-to-end application experience for AI-powered contact centre analytics but also gives us the flexibility to do custom BI analytics and the agility to build proprietary data science models using its rich metadata."
Other announcements relating to machine learning include Amazon Kendra (an enterprise search system for natural language search across multiple silos of information), Amazon CodeGuru (for automating code reviews that identify the most expensive lines of code), Amazon Fraud Detector (the same technology that Amazon uses to detect transaction fraud), Amazon Transcribe Medical (automatic transcription of medical speech), Amazon Augmented Artificial Intelligence (for human review of ML decisions), and new capabilities for Amazon SageMaker (including an ML IDE, elastic Jupyter notebooks, experiment management, a debugger, automated model building, and concept drift detection).
Three new security services were announced: Amazon Detective to simplify security investigations, AWS IAM Access Analyzer to simplify the auditing of access policies, and AWS Nitro Enclaves to isolate highly-sensitive data.
AWS also announced several EC2 changes, including new Arm-based instances, AWS Inferentia instances (for high-performance, low-cost inferencing), AWS Compute Optimizer (to help select the optimum instance types for workloads), AWS Transit Gateway (to improve connectivity between AWS virtual private clouds and customers' data centres), and AWS VPC Ingress Routing to connect virtual private clouds and third-party appliances).
AWS ANZ managing director Paul Migliorini, said "Today's launches are the next iteration of our continued innovation to help solve problems on behalf of our customers. AWS is the most comprehensive and broadly adopted cloud platform, with 175 services that help customers power their infrastructure, become more agile, and lower their operating costs. It provides businesses from startups to large enterprises with the breadth and depth of everything they need to succeed.
"We are entering a formative phase for artificial intelligence and machine learning where we able to bring together more services to help provide accessibility and more choice for customers. Customers want to perform fast analytics on raw data across their data warehouse and data lake, and deal with the explosion in log data, cost-effectively retaining information that might be applied to help run their businesses better. With today's announcements we are helping AWS customers do all of this, and embrace data at scale.
"Customers trust us to deliver security and reliability of AWS to run their most critical applications in the cloud, which is why we are excited to announce new security features such Amazon Nitro Enclaves. This will make it easier for customers to process highly sensitive data by partitioning compute and memory resources within an instance to create an isolated compute environment.
"The announcement of AWS Outposts, which brings native AWS services, infrastructure, and operating models to virtually any datacenter, co-location space, or on-premises facility will help bring further accessibility to organisations based in regional Australia opening up new opportunities to grow and innovate. With our new AWS Wavelength service, we will also be able to bring 5G and cloud together to give our customers the powerful new capability to run cloud services consistently within a few milliseconds of mobile end-users."
Disclosure: The writer holds National Australia Bank shares.