Technology Blogs by SAP
Learn how to extend and personalize SAP applications. Follow the SAP technology blog for insights into SAP BTP, ABAP, SAP Analytics Cloud, SAP HANA, and more.
cancel
Showing results for 
Search instead for 
Did you mean: 
Zora
Advisor
Advisor
Although it’s already been a month since the AWS re:Invent 2019 conference, the customers and analysts are still pouring over hundreds of announcements and launches that were made at the event, trying to discern which ones are the most relevant to their businesses and the future of the cloud market in general.

In my role as a Sr. Cloud Architect with Multi Cloud Technology Office, SAP Global Cloud Services, I followed very closely the conference. In this blog, I'll be sharing some of the highlights from the event – mainly focusing on the most important news from the perspective of SAP customers.

Event highlights


AWS made a series of major announcements in pretty much all areas of IT infrastructure and cloud applications development that clearly demonstrate the company’s relentless investment in new product development and innovation.

The event had a record attendance of more than 60,000 attendees (almost 3 times more than last year), it featured more than 2500 sessions, and had 77 new products, features, & service launches in just 5 days.

In the words of AWS CEO Andy Jassy, their ultimate mission is...

“... to be the infrastructure technology platform underneath all of these enterprises in their transformation strategies and to enable them to be able to invent and to build better customer experience and to help them grow.”

SAP CTO juergen.mueller joined the customers ENGIE and HP on the stage for the session about how companies are transforming their SAP solutions with AWS. Dr. Mueller talked about SAP’s 11-year strategic partnership with AWS and reflected some of the major joint accomplishments of SAP and AWS, including:

  • 5000+ customers worldwide running SAP on AWS

  • SAP Service Cloud Integration with AWS

  • AWS leadership in supporting the largest number of SAP SCP services on their cloud platform

  • Support for super-large SAP HANA High Memory instances (e.g. 18/24TB HANA, 48TB S/4HANA)


Dr. Mueller then goes on to describe SAP's Business Technology Platform strategy for the Intelligent Enterprise in the cloud, which is the key tenant of a joint SAP-AWS business architecture shown below and the future focus of the two companies collaboration:
Source: How ENGIE, HP, and SAP are transforming their SAP solutions (AWS re:Invent 2019)

 

Here are a few selected services and launches of interest to SAP customers:

1. General Availability of EC2 High Memory Instances for SAP HANA with up to 24 TB of RAM



  • EC2 High Memory instances come in 6, 9, 12, 18, and 24 TB memory sizes. They are purpose-built to run large in-memory databases in the cloud and AWS collaborated with SAP to get them optimized and SAP-certified for production deployments of SAP HANA in AWS cloud. The AWS Nitro System technology is used to offload networking, storage, and host management to a dedicated hardware – thus providing applications with direct access to all the processor and memory resources of the underlying server. With EC2 high memory instances, customers can run both SAP HANA and business applications on the same, shared Amazon Virtual Private Cloud (VPC) – which significantly reduces the management overhead associated with complex networking and hybrid architecture.


Key benefits for SAP HANA customers include high scalability and flexibility of scale, simplified management, built-in support for elastic storage capacity (Amazon EBS), and seamless integration with other AWS services – as shown below:
Source: Scale Your SAP HANA workloads on EC2 High Memory instances (AWS re:Invent 2019)

 

2. New hybrid offerings: AWS Outposts, Local Zones, and Wavelength


AWS made three announcements to further boost their hybrid story - General Availability of AWS Outposts, and availability of AWS Local Zone and AWS Wavelength.

According to AWS,

  • AWS Outposts  “is a fully managed service that extends AWS infrastructure, services, APIs, and tools to virtually any customer datacenter, co-location space, or on-premises facility for a truly consistent hybrid experience. AWS Outposts is ideal for workloads that need low latency access to on-premises applications or systems, local data processing, or for local data storage needs."

  • AWS Local Zone “is a new type of AWS infrastructure deployment that places compute, storage, and other select services closer to customers, giving developers the ability to run applications that require single-digit millisecond latencies to end-users in Los Angeles, CA (Los Angeles is the only location supported with AWS Local Zone as of now).”

  • AWS Wavelength “enables developers to build applications that deliver single-digit millisecond latencies (e.g. game and live video streaming, machine learning inference at the edge, AR/VR) to mobile devices and end-users. AWS developers can deploy their applications to Wavelength Zones, so that application traffic can reach application servers running in Wavelength Zones without leaving the mobile provider’s network.”


These capabilities have significantly improved the AWS hybrid story: Whether customers want a low latency experience on-prem with Outposts, or lowest-latency in the public cloud with Local Zones, or in the core carrier network with Wavelength - AWS can now deliver the complete package.

3. Many new AI/ML offerings


AWS added several new offerings to their already impressive  AI/ML portfolio:

In the ML infrastructure area, Amazon announced the availability of EC2 Inf1 instances based on their custom silicon for machine learning inference chip called “Inferentia”, that was pre-announced last year. According to AWS,

  • With Amazon EC2 Inf1 instances, customers receive the highest performance and lowest cost for machine learning inference in the cloud. Amazon EC2 Inf1 instances deliver up to 3x higher throughput and up to 40% lower cost per inference than Amazon EC2 G4 instances, which were already the lowest cost instance for machine learning inference available in the cloud.”  - For more details, see:  EC2 Inf1 instance announcement.


In the ML frameworks area, Amazon introduced the SageMaker Studio which was touted as the first comprehensive IDE for machine learning.

  • The SageMaker Studio makes it easier for enterprise to develop ML services.  Organizations do not need to hire anymore an army of data scientists and ML-savvy developers who are very hard to attract and retain. Also, their developers’ productivity will get a boost from SageMaker Studio by having a central place to access all the necessary infrastructure components like GPUs, FPGAs and ASICs, as well as toolsets and services for developing, training, running, monitoring and debugging ML services.


In the ML services area, Amazon launched a couple of new tools that are targeting people who are not ML-savvy developers:

  • Amazon CodeGuru is machine learning service for automated code review and application performance profiling that helps software developers automate code reviews and identify an application’s most expensive lines of code.

  • Amazon Augmented AI (A2I) makes it easy to build workflows required for human review of ML predictions. It helps ML developers validate machine learning predictions through human confirmation.


Amazon hopes that these services will have a great appeal to smaller enterprises which often do not have enough skilled ML developers and can benefit greatly from using the pre-packaged tools and services for easier development.

4. Cost and performance optimization-related announcements


With cloud cost optimizations consistently ranked as the #1 topic among customers, AWS wanted to demonstrate their continuous innovation in this area by shipping several new optimization features, such as:

  • AWS Compute Optimizer: Compute Optimizer leverages machine learning to analyze customers utilization data and generates recommendations to help customers reduce costs and improve performance by choosing the optimal EC2 instance types. It also delivers recommendations for instances that are part of auto scaling groups with a fixed group size and infers how the workload would have performed on various hardware platforms (e.g. different Amazon EC2 instances types) and offers recommendations. There is no cost for the service, customers only need to pay for the compute resources needed to run their applications and for Amazon CloudWatch monitoring fees.

  • Accelerated Site-to-Site VPN uses AWS Global Accelerator to improve the performance of VPN connections. The traffic from the customer gateway device is routed through the closest AWS Edge location and traverses the AWS global network to reach the customer’s VPN endpoint in AWS.


5. New announcements around container and Kubernetes-based services


Amazon launched several new Kubernetes/container-based computing services to meet a growing demand from enterprises for flexible and cost-efficient infrastructure that works both on-premise and cloud:


  • AWS Fargate (a serverless computing option for containers) now supports Amazon EKS, a hosted Kubernetes service – which greatly simplifies provisioning and management of compute clusters running containers. AWS Fargate eliminates the need for customers to create or manage EC2 instances in the EKS clusters – thus making it much easier for customers with no expertise in Kubernetes operations to run cost-optimized and highly-available clusters. Using Fargate, customers do not have to worry about patching, scaling, or securing Kubernetes clusters. And thanks to the ability to define and pay for resources at the pod-level, customers can see the cost of each pod – which makes it much easier to right-size resource utilization for each application.




  • Customers can further save by utilizing the newly announced AWS Fargate Spot feature which offers a 70% discount on compute cost, in exchange for giving AWS permission to take back the compute and delete your instances whenever they need to.




  • AWS also announced a new auto-scaling capability, the ECS Cluster Auto Scaling that provides customers with more control of when and how Amazon ECS clusters scale-in and scale-out.




These are the top 2019 re:Invent announcements that you need to be aware of. Want to learn more? Checkout the links below:


  1. To see all 77 announcements made at the event and to read about them in more detail – click here: 2019 re:Invent Announcement Summary Page




  2. Check the table below for a list of all SAP-related sessions at the event. For the complete list of all sessions, see  A list of all sessions with a PDF available.