Oracle launched its previously on-premises database machine, Exadata X8M, to its public cloud.

The Oracle Exadata Cloud Service X8M on Oracle Cloud Infrastructure uses the new X8M architecture. It is engineered for mixed workloads, and consolidating transaction processing and data warehousing into the same database, on the same cloud service, the vendor says.  

Oracle, per usual, also made it a point to one-up its rival Amazon Web Services (AWS), citing how Exadata Cloud Service X8M can provide database input/output operations up to 5,000% faster than AWS Relational Database Service (RDS), besting Amazon RDS’ CPU scaling performance by 250% and storage performance by 200%.

Similar to its on-premises counterpart, the Exadata Cloud Service X8M allows users to start at a quarter rack, and it enables them to independently add resources including both compute and storage, while paying only for what they use with per second billing, wrote Carl Olofson, research VP of data management software at IDC, in a blog post. 

“Exadata X8M’s performance combined with granular elasticity lowers overall cloud costs as customers complete tasks faster with smaller configurations than competitive database cloud service offerings,” Olofson said.

In his post, Olofson recalls a time when some database management systems (DBMSs) were co-designed with hardware platforms specifically architected to run them. These DBMSs were also designed for the most extreme and demanding database workloads, he says. 

“As hardware costs dropped and performance improved, however, most database users opted for DBMSs that ran on standard reduced instruction set compute (RISC) or x86 servers with network-attached storage (NAS) or storage area network (SAN) storage systems, and the database machines were relegated to the job of managing extremely large enterprise data warehouses (EDWs),” Olofson explained.

“In time, most of these database machines faded into obscurity,” he added. “Oracle’s Exadata Database Machine is an obvious exception to this trend — still running strong with 15% growth as reported by Oracle in Q1 and continuous innovation over 10 years with demonstrable superiority over generic x86 servers running Oracle Database. Oracle Database is also not a specialized database, as it can run on other vendors’ servers and clouds—just not as well.”

Oracle first unwrapped the Exadata X8M platform during its annual OpenWorld event, introducing what the vendor dubbed the world’s fastest database machine. The platform brings together Intel Optane persistent memory and 100 gigabit remote direct memory access (RDMA) over converged Ethernet (RoCE) to remove storage bottlenecks and increase performance for demanding workloads such as online transaction processing (OLTP), analytics, IoT, fraud detection, and high frequency trading. 

Exadata Cloud Service X8M is based on Oracle Real Application Clusters (RAC), and optimizes communications between the compute nodes and storage cells using remote direct memory access (RDMA). This, Olofson says, allows the compute node to read and write data directly from or to the storage cell, “greatly streamlining communication.”

It also features a hot-warm-cold approach to storing data. Cold data is stored on spinning disks, warm data is stored as flash memory that is directly managed for optimized reads and writes, and hot data is divided between main memory on the database server and persistent memory in the storage server depending on its volatility. 

According to Oracle, the new service’s software-defined flexibility enables up to 25 petabyte’s (PB) database capacity. “Running Oracle Database on OCI not only means a more flexible Exadata X8M deployment, but it also means the option of running the database under Oracle Autonomous Database, which self-tunes and self-heals driven by artificial intelligence (AI) and continuous machine learning (ML) process, is also available,” Olofson writes.

Expanding its cloud services has been a high priority for Oracle due to customer demands.

“Customers have told us that to run critical systems of record in the cloud, they need to run workloads across fully independent cloud regions for disaster recovery purposes,” said Oracle’s director of product management, Andrew Reichman earlier this year. 

Still, most of Oracle’s cloud customers are simply moving their on-premises Oracle workloads to Oracle’s Cloud, as opposed to net new customers. “Clearly, they’ve been beefing up their infrastructure play,” said Gartner analyst Sid Nag in an earlier interview with SDxCentral. “But engineering doesn’t win the day. You can have great engineering infrastructure, but how are you going to attract customers? The key piece that will drive the market share catchup game would be getting more net new clients.”