Amazon's Cloud Service and Data Analytics Expand
Wed, April 21, 2021

Amazon's Cloud Service and Data Analytics Expand

Amazon wants to provide one-stop shopping for all data management and analytics needs on AWS / Photo Credit: Piotr Swat (via Shutterstock)


Amazon Web Services (AWS) wants all organizations of all sizes to run databases in the cloud, reported Sean Michael Kerner of TechTarget, a marketing company. At the AWS re:invent 2019 in Las Vegas, the cloud giant stated its Amazon cloud database strategy, which leverages on using multiple purpose-built offerings for various use cases. Carl Olofson, an analyst at IDC said, “Quite simply, Amazon is looking to provide one-stop shopping for all data management and analytics needs on AWS.” 

AWS CEO Andy Jassy said that many companies use relational databases for each of their workloads, “and the day of customers doing that has come and gone.”In his perspective, there is too much data, complexity, and cost involved in utilizing a relational database for all workloads, sparking demand for purpose-built databases. For example, Lyft has millions of drivers and geolocation services, but this information is not suitable for a relational database, Jassy explained. 

For cases similar to Lyft’s, a fast, low-latency key value store is needed. This is why AWS has the DynamoDB database. An in-memory database is crucial for workloads that “require sub-microsecond latency.” This is where ElastiCache comes in. If you’re looking to connect the dots across multiple large data sets, then a graph database is a great option, which is what the Amazon Neptune service offers. Alternatively, DocumentDB is a document database, which you can use if you are working with documents and JSON. 

AWS also offers its Apache-Managed Cassandra database to provide users with single-digit millisecond latency without worrying about managing data clusters. AWS also announced its Redshift RA3. Jassy stated that as users consume the storage space of a Redshift local instance, the RA3 service will transfer the “less frequently accessed data over to S3” cloud storage buckets.  

Lastly, Jassy showcased the new Advanced Query Accelerator (AQUA) for Redshift. It provides an innovative way to conduct hardware-accelerated cache to bolster query performance. AWS has developed a “high-speed cache architecture” along with S3 to “scale out in parallel to many different nodes." The nodes have custom-designed AWS processors to boost the speed of operations. Jassy added, “This makes your processing so much faster that you can actually do the compute on the raw data without having to move it.”