Introduction

 

Exadata Cloud@Customer X10M introduces cloud automation and consumption pricing, akin to the Exadata Database Service and fully managed Autonomous Database in Oracle Cloud Infrastructure (OCI), directly into an organization’s on-premises data centers. This service, owned and remotely operated by Oracle, enables organizations to seamlessly leverage cloud resources for critical applications, integrate with existing on-premises systems and data sources, and comply with data residency requirements.

 

 

High performance architecture

 

Exadata Cloud@Customer X10M utilizes a distinctive scale-out design, incorporating fourth-generation AMD EPYC processors in both database servers and intelligent storage servers. This configuration offers an increased number of processing cores and enhanced per-core performance compared to earlier platforms. The intelligent storage servers play a crucial role in expediting SQL processing by offloading it from database servers, facilitating faster SQL I/O. Additionally, the Exadata RDMA Memory contributes to reducing SQL read I/O latency to less than 17 microseconds.

 

 

High transactional performance

Exadata’s distinctive RDMA-enabled scale-out architecture enhances the performance of OLTP applications. The storage servers employ the Exadata RDMA Memory Data Accelerator, coupled with a large NVMe flash cache, to automatically cache frequently accessed data at extremely low latency. In the case of Exadata Cloud@Customer X10M, quarter-rack configurations begin with 380 available processor cores, 5.6 million SQL low-latency read I/O operations per second, and nearly 2 TB/sec of memory bandwidth. These configurations are scalable to meet the transaction processing demands of a wide range of organizations.

 

Accelerated analytics and machine learning

Analytical tasks and in-database machine learning (ML) processes achieve accelerated performance through the elimination of I/O operations and the delegation of analytic SQL queries and ML algorithms to storage servers. Exadata Cloud@Customer X10M excels in analytics with a starting point of 135 GB/sec of scan throughput, 190 TB of uncompressed database capacity, and the capability to accommodate up to 1.9 PB of data warehouses using Oracle Hybrid Columnar Compression in a quarter rack. Its scalable architecture supports up to 2,880 GB/sec of scan throughput, 4 PB of uncompressed databases, and a substantial 40 PB of Hybrid Columnar Compression compressed data warehouse capacity, making it ideal for data-intensive workloads.

 

Independent database compute and storage scaling

The database servers deliver 190 functional processor cores and a capacity of up to 2,800 GB of memory, while storage servers contribute 64 processor cores and 80 TB of usable storage. Flexibility is maintained as database and storage resources can be independently scaled by adding individual servers. The initial rack accommodates up to 16 servers, and across multiple racks, there is potential for up to 32 database servers and 64 storage servers. The software stack undergoes optimizations that guarantee linear scalability in database performance when operating on Exadata X10M servers, characterized by a high core count and substantial memory capacity.

 

Greater database consolidation

Customers can achieve enhanced management efficiency and cost savings by consolidating a greater number of databases within a single Exadata Cloud@Customer X10M compared to the previous generation platform. This consolidation capability enables organizations to concurrently operate the automated Exadata Database Service and fully managed Autonomous Database on the same Exadata Cloud@Customer infrastructure, facilitating the efficient consolidation of various workloads.

 

 

Summary


Oracle’s holistic approach empowers IT leaders by offering control over database services while benefiting from extensive automation. The platform also extends its capabilities with Exadata Cloud@Customer X10M, bridging the gap between on-premises data centers and cloud automation. With a focus on simplified management, cost-effectiveness, and adaptability, Oracle’s Autonomous Database emerges as the forefront of the future of data management.

https://www.oracle.com/autonomous-database/

https://blogs.oracle.com/database/post/a-guide-to-why-and-how-to-consolidate-databases

 

(Gupta, Divit, n.d.)