Takeaways from AWS re:Invent 2022
In December 2022, I had the opportunity to attend the AWS re:Invent summit at Las Vegas and observed announcements on new products and functions of the AWS cloud ecosystem. We will describe some of the announcements observed during my time at the summit.
Overview of Product Announcements
AWS announced developer tools augmented with AI, with the goal to improve developers’ productivity. DevOps is viewed as a data science problem, with logs used as data source for analysis. AI will aide code development, review, and operations. There are 3 tools mentioned at the conference session on AI for developers. First is AWS CodeWhisperer, which is an ML-powered coding companion that generates code recommendations based on users' comments in natural language and their code in the integrated development environment (IDE). Next is Amazon CodeGuru, a developer tool that provides intelligent recommendations to improve code quality and identify an application’s most expensive computation and inefficient lines of code. Lastly, there was a showcase of Amazon DevOps Guru, which uses ML to detect abnormal operating patterns such as analyzing streams of disparate data and monitor relevant metrics to restore normal operating patterns.
To explore data, we need the right tools, integration, governance, and visualization. Tools like Amazon Aurora for MySQL and PostgreSQL, and integration with Amazon RedShift provide better user experience for data and analytics workflow. AWS has the goal to remove ETL processes and move towards a zero ETL future. A fully managed zero ETL combines AWS’s functions. For instance, users can choose Aurora tables with Redshift analytics function. Aurora database can be replicated to the same Redshift instance in near real-time. This allows the infrastructure to scale up or down with no up-front infrastructure to manage. Data analysis can also be conducted with Apache Spark, which runs about 3 times faster on AWS. Today, Spark must move to Redshift from S3. Going forward, Redshift is integrated with Spark and Juypter, with no need to manage any connector or to manually move data around. There was also an announcement on AWS Clean Rooms, which help organizations collaborate with external business partners on secured datasets, without revealing the underlying data. Celent covered clean rooms as part of the confidential computing topic in this report, “Securing Insurance Data: Confidential Computing and Data Lineage Use Case”.
On governance, there needs to have a balance between having the right controls and access. Data governance across an organization is complicated, with communication between different functions and teams. Amazon DataZone was introduced to provide a way to discover, share, and manage governance controls in data. Essentially, data management involves people in the organization and the data within. An ideal scenario is to have the right metadata tagging for the data catalogue, with right access/controls and analytics integration. Celent shares this similar perspective in the blog, “At the Heart of Design - Adopting Machine Learning with a Data-Focus Design”. Meta-data will provide a shared knowledge base and enable process like MLOps to be inserted into the wider organization functions. Metadata will provide the common library for both the machine and human stakeholders to utilize machine learning to meet business objectives.
For security offerings, AWS showcase Amazon GuardDuty, a threat detection service that continuously monitors users’ AWS accounts and workloads for malicious activity and delivers detailed security findings for visibility and remediation. It also offers Elastic Kubernetes Service (EKS) protection, allowing users to run Kubernetes on AWS securely through detecting threats on Kubernetes clusters. Amazon Security Lake centralized security data from cloud, on-premises, and custom sources into a purpose-built data lake into users’ account. Security Lake provide Open Cybersecurity Schema Framework (OCSF) support, with connections to vendors such as Cisco and Palo Alto Networks.
And for simulation tools, AWS SimSpace Weaver was announced, which is a new compute service to run real-time spatial simulations in the cloud and at scale. Organizations can run simulations that are rare, dangerous, and expensive to test in the real world. For instance, scenarios such as the impact of natural disasters on a city can be tested as part of an organization's impact assessment. SimSpace Weaver can be run on Amazon Elastic Cloud Compute Cloud (EC2) instances.
AWS APAC Market Update
Closer to my base market, I was also presented with an update for AWS financial services in APAC. Three main items describe the changes at APAC, namely cloud migration across the region, migration to launch new services, and continued debate on build versus buy. Applications migration on the cloud are now focused on short release cycle, developers’ productivity, and cost efficiency. Examples include National Australia Bank which has migrated 70% of its application to the cloud. Cloud migration for the purpose of launching new services contribute to organizations’ success with digital transformation and innovation strategy, with success metrics using NPS scores, customer experience, and new revenue stream creation. An example is Axis Bank utilizing AWS services to provide digital experiences for its customers. The build versus buy perspective is that new players will not build whole but to train builders/talent internally and to buy externally at the same time. SaaS was preferred for functions with a lower priority and to build internally for important functions. AWS observed that most FIs in the region are conducting training programs to train talents.
AWS offerings at also aimed at improving the core systems of FIs, such as core systems’ optimization, upgrade, and replacement. In insurance, the migration of policy administration systems to the cloud provides better configurability and towards a shared data ecosystem. This allows external data and unstructured data to be utilized and help FIs build applications such as embedded or parametric insurance.
During the Q&A session between analysts and Adam Selipsky, CEO of AWS, he provided a good overview of where AWS is heading towards. AWS core foundational services like EC2 and data analytics are still part of the company’s key strategy for growth. The emphasis of the summit is to highlight their commitment to go deeper into business verticals and to accomplish that by building integration between services. Services that will provide for both higher level and lower level (such as foundational services). Organizations need not replatform everything but to identify what is most crucial to the business first. Customers’ strategy should drive the services they use, and adopting AWS for cloud integration services, management, and data governance. The belief is that cloud offers flexibility, have resources on demand, and help organizations be more agile. AWS aims an important part of businesses, and to have more CEOs’ conversations over technical discussions with developers, leading to a customer first approach to determine the right strategy. And in terms of talent development, Adam imparted this phrase of being “strategically patient and tactically impatient”. People are hired as resources to accomplish the long-term objectives, and AWS will help industries by investing $25 million by 2025 to train people to be cloud ready.
To learn more, Celent tracks this market and has research addressing it (list of recent reports here).If you would like to find out more, please feel free to get in touch with me.
Below are related reports contributed by Celent on this topic:
At the Heart of Design - Adopting Machine Learning with a Data-Focus Design
Securing Insurance Data: Confidential Computing and Data Lineage Use Case
The Growth of Confidential Computing for Secured Data Sharing on the Cloud