AWS reInvent 2022: The Most Interesting

AWS Vice President Jeff Barr and a select group of AWS Developers personally selected some of the most interesting product and service launches to debut at AWS re:Invent 2022. Our AWS experts compiled the most exciting announcements at AWS re:Invent 2022.

In this article: 

AWS Glue 4.0

AWS Glue is a managed Extract, Transform, Load (ETL) service that can process data stored in S3 or DynamoDB and convert it to other formats or schemas for better use in other services such as Athena. AWS Glue is for people who have too much data to process. Maybe you have a whole fleet of servers, and each of them outputs log files. You load this data into S3 for easy storage, but there is a lot of it and it needs to be processed first before it can be analyzed with Athena. Perhaps you are only interested in a few columns of data and want to delete the rest. AWS Glue treats data as a utility. Once you've set up your robot, you can import the data stored in S3 into your data directory, the same directory that Athena uses to run queries. You can then edit this data to remove unwanted columns or convert formats.

Glue 4.0 includes newer upgrades on Spark, Python, and Scala. The users get Spark 3.3, Python 3.1, and performance-optimized Spark 3.0, which has multifold performance or open-source Spark. AWS users can also use distributed Pandas API to run Pandas workloads. Also, AWS announced a new highly-performed Redshift connector that is 10x faster on TPC-DS benchmarking. It gives users access to their data and data warehouse quickly.

AWS Glue for Ray

Talking about Python you get framework support, like NumPy , SciPy,  Pandas, and many more. You also can bring your own libraries. Now you can build applications with Ray. The AWS customers can distribute their workload, by doing some basic decorators. All you need to do is take your Python code and do the decorator like ray.something and specify the parallelism that you need. And it will help you to scale your single-node Python and get the similar capabilities that you have for Apache Spark, but now you can use Python end-to-end. 

Amazon Transcribe – Real-Time Analytics During Live Calls

Amazon Translate is a service that uses machine learning to translate text between different languages. The service is delivered through an application programming interface that allows companies to quickly connect it to their chatbots and customer support systems. Real-time call analytics provides developers with APIs to accurately transcribe calls in real-time, as well as identify problems and customer sentiment in real-time.

AWS introduced real-time call analytics in Amazon Transcribe Call Analytics to help customer support agents process real-time calls faster. Real-time call analytics enables software engineers to develop APIs to accurately transcribe calls in real-time, as well as identify problems and customer sentiment in real-time. Transcribe Call Analytics uses state-of-the-art machine learning to automatically evaluate thousands of pending calls and identify customer service issues such as repeated requests to speak to a manager or unsubscribe.

Did you know?

Develop only with best vendors

Now, Geniusee team provides you with even better results in on-demand cloud computing platforms and APIs.

Check it 

New AWS Local Zones

Local Zones are AWS infrastructure that places computing power, data storage, databases, and other services closer to major cities and industrial centers. In this way, AWS customers will be able to develop applications requiring minimal latency, less than one-hundredth of a second, close to end users. This entails seamless access to applications such as real-time gaming, multimedia, entertainment content creation, live streaming, assimilation, augmented and virtual reality applications, etc.

Just the other day, 4 new local zones appeared - Buenosaires (Argentina), Copenhagen (Denmark), Helsinki (Finland), and Muscat (Oman). In total, 27 new local zones in 22 countries have been added this year. The expansion of local zones is necessary for latency-sensitive workloads, for example, advertisement, game tech, and other cases when it is important that the transfer date passes with minimal delays in real-time.

Open Search Serverless

The Open Search (OS) is a search engine, just like Google or Bing, but for your own data. It will index your data and make it searchable through the HTTP interface. You can download the OS and install it on any computer you like. Install it on your local machine, set up an EC2 instance on AWS and install it there, or let AWS help you get it up and running with AOS.

Now there is a new deployment option - Open Search Serverless, along with dashboards. Now it does not offer a choice and configurations of CPU, memory, etc. The architecture of Open Search Serverless is reminiscent of Aurora. Here, the “data storage” layers (they are made on the basis of S3) and “compute” layers are completely separated. Moreover, “compute” is separate for writing, reading, and searching. Accordingly, these layers can be scaled independently of each other. Also in the console, you can select the maximum limit separately for writing and for reading above which Open Search will not be scaled.

Subscribe to our news

Subscribe

X

Thank you for Subscription!

AWS SimSpace Weave

AWS SimSpace Weave is a tool for emulating 3D spaces and cities. With this service, you can solve 3 problems:

1. Real-Time Simulation (simulation on real cities, on real maps, taking into account real traffic, real routing capabilities, and connectivity). This can be useful for Supply Chain logistics, which are not land-based trade communications (by water, by air).

2. Simulation of Synthetic Environments (designing trade deliveries with possible geographical changes - the channel is blocked, the territory for trade is closed)

3. Simulation of the Virtual World (first of all, this is Metaverse and various game simulations. This helps to build a game with good game mechanics.)

DataZone

Unfortunately, it is impossible to try this program, tentatively it will be possible at the beginning of 2023. This service is needed so that the data consumers that analyze this data can meet with the creators of the data. This is a data catalog that makes data visible with context for all your users to find and understand data quickly and more easily. You can catalog any data type across the organization, in the cloud, on-premises, or in SaaS applications with rich metadata and business context. DataZone also can be used for governed data sharing. It can help to automate workflows to facilitate sharing of data more securely between producers and consumers to make sure that the right data is accessed by the right users for the right purpose. Additionally, you can use the DataZone AWS tool to efficiently audit who is using which datasets for what business use case, and monitor usage and costs across projects and lines of business. 

Amazon Security Lake

Amazon Security Lake is a security log aggregator that helps you automatically centralize your security data in a few steps. There is an S3 Bucket under the hood, in which you can add logs from other sources.

With the help of Amazon Security Lake you can:

• Use your preferred tools with centralized data. Makes it easy for teams across your organization to access data with their preferred analytics tools, without data duplication.
• Analyze data using your custom analytics. Gives security teams broader visibility to investigate and respond to suspicious activities using your custom analytics models.
• Support on-demand analysis of petabyte-scale data. Builds a petabyte-scale data lake to support security investigations and rapid incident response scenarios. 
• Analyze multiple years of security data quickly. Centralizes years of cloud, on-premises, and custom data in your S3 Buckets, and use your preferred tools for security analytics. 

AWS CleanRooms

AWS CleanRooms is a data management tool that helps companies and their business partners securely analyze and collaborate on datasets without sharing or revealing their underlying data. At the same time, you do not have to disclose your basic data to other collaborators. In the AWS Console or using the API, customers can select collaboration partners and desired datasets, and then set limits for each member. However, you have the option not to work outside of the AWS environment and not to upload data to other platforms. Multi-party collaborations allow up to 5 parties in a single collaboration, extracting insights from multiple companies. 

AWS Nitro System v5

AWS Nitro is a combination of purpose-built hardware and software designed to maximize performance and security. At the same time, Nitro includes several components at once, among which the main ones are a DPU, a dedicated security chip with hardware RoT, and a lightweight hypervisor. This allows you to offload the central processor, since now access to the network and resources, security issues, and more depend on dedicated hardware controllers. And the hypervisor manages only the allocation of memory and computing resources.

In AWS Nitro System v5, a lot of emphases was placed on bandwidth, and as a result, packets per second increased by 60%, latency decreased by 30%, and battery energy consumption per watt decreased by 40%. All this in turn leads to a reduction in costs by 30-40%.

AWS Lambda SnapStar

AWS Lambda is a dedicated tool that helps you activate any code for any application without having to administer it. Any code will be executed based on events that will occur in AWS services. In simple words: you add your script, which must be executed in AWS Lambda, and set a trigger or event that will run this code.

With the advent of AWS Lambda SnapStar, Java developers will be able to use Lambda to write requests. Lambda can be started via cold start or warm start. A cold start is the first launch of Lambda or a launch after a long time, where you need to create an execution environment, download the code and create a runtime for the programming language. SnapStar allows you to reduce the execution time of INIT functions. This is because when a new version of Lambda functions is published, the Lambda service automatically initializes it and takes a snapshot of memory and disk space. And as a result, it will take less time to run Java code.

Amazon Omics

Amazon Omics is a tool that simplifies the storage, transfer, and analysis of genomic and other multi-omic data for accurate healthcare results. It helps researchers and clinicians focus on getting deeper insights from existing data by simplifying and accelerating the storage and analysis of multi-omics information.

You can use Amazon Omics storage to bring together petabytes of omics data in one place, enabling scientific analysis and population-scale research. This tool can be used to automate the integration and scaling of computing infrastructure to reduce the time for infrastructure management and increase the time for research.

Conclusion

In 2022, 70 new products and services were presented at the AWS reInvert conference, but we chose the most interesting ones. Our AWS-certified development team stays on top of the technology market to provide the latest and most innovative solutions for your business. If you have any questions about AWS solutions for your project, please fill out the contact form and get a free consultation with our specialists.