Oracle Database services on OCI are the linchpins of Oracle Data Platform, a comprehensive, open, and integrated environment that helps organizations get the most value out of their data with consistent governance, security, and performance with a minimal amount of integration. To do this, the services that make up the data platform must work with many types of data and do so with high performance for a multiplicity of workloads. However, database workloads in the public cloud are highly varied. Many are simple and require only fairly prosaic configurations. Some of their concurrent user access rates are relatively low, and their database operations are simple. Other workloads are extreme, with large numbers of users accessing the database at the same time, all expecting immediate responses. When they become frustrated, database administrators tend to respond by adding more servers with more memory and IO bandwidth to the database cluster, but this approach pushes the cloud infrastructure bill ever higher.
Oracle Database tackles some of the most demanding database workloads on the planet. It provides built-in support for the variety of data and workloads that support the modern enterprise and fast growing SMBs. Oracle Exadata adds performance on top of versatility and is designed to ensure that business-critical workloads, whether they are transaction- or analytics-oriented, can always be served quickly and efficiently. In the Oracle Cloud (Oracle Cloud Infrastructure, or OCI), the hardware enables independent scaling at the compute and storage layers, which means that Oracle Database can deliver much better performance on OCI than on any other cloud service.
The storage layer for Autonomous Database and Exadata Database Service is already ideal. The underlying Exadata system that supports both services segregates the data based on demand into hot (memory), warm (flash) and cold (disk) storage automatically. Data transfers are accomplished using direct access to the storage memory with remote direct memory access (RDMA) over a private RoCE fiber network. But on the compute side, Oracle users keep pushing the limits. Up to now, with ever more transactional users, and ever more complex queries on the analytic side, the Exadata compute nodes could still use some tweaking to optimize performance.
To address this need, Oracle has made available Oracle Database services in OCI, including Oracle Autonomous Database, on its latest system, Exadata Cloud Infrastructure X9M. One key advantage to customers of this offering is that compute nodes are now powered by AMD EPYC™ processors. AMD EPYC enables faster, smoother operations and more effective application of compute power to the problems solved in the database. It also enables smaller database systems in cases where it makes sense, doing the same job as larger ones, since the new system delivers 2.5 times more database processing cores per configuration.
Exadata X9M systems in OCI enable Oracle Database services to scale more effectively without boosting user cost by scaling up service consumption during periods of peak demand and back down when demand subsides. This serves not only the largest and most demanding workloads, but also more modest needs, since, especially with Oracle Autonomous Database, the technology can smoothly and transparently adjust to the exact needs of the user without requiring sudden increases and decreases in the number of database nodes (and the downtime associated with these configuration changes) that one experiences with other cloud DBMSs.
AMD EPYC processors are becoming increasingly common as the go-to processors for the cloud. Their ability to concentrate more power on any workload (supporting up to 64 cores per socket) means better performance for Oracle Database users in OCI. This is especially true for transactional workloads where more cores packed together means greater throughput for large volumes of parallel access. With today's increasingly complex, logic-driven database updates, it’s important to be able to execute multiple embedded SELECT clauses, and update multiple table rows in a single operation, within a very thin slice of time. In the OLTP world, throughput is everything. Also, as database analytic problems become more compute-intensive, especially in terms of complex queries, the ability to pack more processor punch into every database operation offers a clearly needed boost. AMD EPYC processors help deliver the necessary compute power to accelerate Oracle Database performance.
Oracle Database users have a variety of options in terms of database deployment. Some remain in the datacenter, either on Exadata or on system configurations that they choose and manage. Others run in the public cloud on various cloud services. Those cloud services can't be optimized for Oracle Database, however, so they tend to offer substandard performance at higher cost than deployments either in the datacenter or on OCI. Oracle Database services on Exadata X9M in OCI seem to be the best overall option for most workloads, both from a performance and cost standpoint. And, with AMD EPYC processors at the core of the new Exadata database servers, the value proposition for adopting Oracle Data Platform leveraging Autonomous Database and Exadata Database Service in OCI just became stronger.
Users of Oracle Database on other cloud services should consider the following:
Carl Olofson has performed research and analysis for IDC since 1997, and manages IDC’s Database Management Software service, as well as supporting the Data Integration Software and Data Streaming Pipelines services. Mr. Olofson’s research involves following sales and technical developments in the structured data management (SDM) software markets. One key market is the database management systems (DBMS) software market, which includes non-schematic database management systems, data lake managers, navigational database management systems, low code database management systems, and memory-optimized shared data managers. Also covered is the database administration and development software market. Mr. Olofson also contributes to Big Data research and provides specialized coverage of Hadoop and other Big Data technologies. Mr. Olofson advises clients on market and technology directions as well as performing supply and demand-side primary research to size, forecast, and segment the database and related software markets.
Previous Post