Issue link: https://resources.awscloud.com/i/1496553
©2020 IDC #US46773920 4 Data Management in the Cloud Some data operations are occasional and project based, whereas others are ongoing and require continuous availability. In the datacenter, one provisions both the same way, allocating systems in the former case that are used only from time to time and others that are scaled to the "high-water mark" of usage when data volumes are high but run at 20–40% capacity the rest of the time. In the cloud, by contrast, resources can be dynamically allocated on an as-needed basis and can scale up and down as required, ensuring that no money is "left on the table." Also, with managed services in the cloud, the staff is relieved of such tasks as patch application, systems maintenance, and database backups. Beyond all this, the cloud offers the opportunity to introduce new data management technologies and address new workloads and integrate them with existing systems in an additive way, without the complexity that normally accompanies such projects when done on premises. Using the Cloud Platform to Bring Differing Data Models Together Moving to the cloud should be a way not only of running existing workloads more efficiently and affordably but also of deploying new database workloads and taking advantage of capabilities offered by the cloud platform itself to switch data management paradigms. Such paradigm switches might include moving some relational tables to documents. The cloud platform should also support adding new data management paradigms (such as are needed for streaming data processing or data graph analysis) and blend the various database operations in an environment that is managed altogether by a single cloud platform provider. AWS Cloud-Native DBMS Options Amazon Web Services offers a broad range of database management technologies that, together, address all the requirements described previously. The aim is to provide a transformative move that AWS calls "refactoring." Refactoring the Databases To really get the benefit of database cloud deployment, one must use DBMS technology built into the cloud environment with tools that enable faster innovation. Whether your need is for relational, document, graph, or other database functionality, the choice should involve technologies that enjoy cloud platform support and the commonality of open source and that offer a net reduction in license and maintenance costs. This means more than deploying an on-premises DBMS in the cloud (AWS calls this rehosting) or moving to a managed version of the same (for AWS, this is replatforming); it means choosing a DBMS technology that is built for the cloud platform and is operationally integrated with cloud services to deliver maximum efficiency and flexibility. One could simply replatform by moving data to a cloud managed form of the DBMS currently in use or similar to it, such as MySQL or MongoDB. But if new workloads or functionality are envisioned, perhaps refactoring is in order. Refactored data is not merely moved but transformed into a format and mode of operation that serves the new workloads made possible in the cloud. AWS DBMSs are offered in a single managed environment and are optimized to take full advantage of the operational characteristics and facilities of the AWS cloud platform. These technologies are offered with APIs that support the APIs of commonly used DBMSs in the relational and NoSQL realms, which means that developers should have no problem understanding how to use them to maximum advantage.