Oracle has announced the general availability of Exadata Exascale, an intelligent data architecture for the cloud that improves performance for Oracle Database workloads—including artificial intelligence vector processing, analytics and transactions—at scale. The new service reduces costs for organisations with built-in performance, reliability and security capabilities, Oracle said in a press release.
It is now available with Exadata Database Service on Exascale Infrastructure and Oracle Database 23ai on Oracle Cloud Infrastructure and will be available in Exadata Cloud@Customer, OCI Dedicated Region and multicloud environments in the future, the company said.
“By reducing infrastructure costs by up to 95%, Exadata Exascale makes it possible for small workloads and small businesses to leverage the benefits of Oracle Exadata for Oracle Databases in the cloud,” said Kothanda Umamageswaran, senior vice president, exadata and scale-out technologies, Oracle.
The data architecture consists of virtualised database-optimised infrastructure on pools of shared compute and storage, which enables hyper-elastic scaling and pay-per-use. Users can specify the number of database server ECPUs and storage capacity they need, and database is spread across pooled storage servers, eliminating the need for dedicated database and storage servers. This reduces the cost of entry-level infrastructure and enables scaling of resources.
According to Oracle, the RDMA-capable storage cloud distributes databases across available storage servers and uses intelligent scan to make CPU cores available to speed up any database query. In addition, data is replicated on three different storage servers to enhance fault tolerance.
Exascale's AI Smart Scan helps offload data and compute-intensive AI Vector Search operations to the storage cloud and, along with System Software 24ai, can run vector search operations faster, allowing users to run concurrent AI vector searches in multi-user environments.
Communication between servers enables database scaling across the Exascale Virtual Machine clusters, and low-latency OLTP IO completes critical transactions and supports more concurrent users. According to Oracle, Exadata Exascale delivers more throughput and lower latency than other database cloud services.
The system also automatically offloads data-intensive SQL queries to the storage cloud, enabling throughput scaling for analytics. Automatic columnarisation converts data into in-memory columnar format that uses flash caches in the storage cloud to increase capability and performance.
Users can create full copies or thin clones using the storage cloud and its redirect-on-write technology. Advanced snapshot capabilities allow the creation of space-efficient clones of pluggable or container databases using read-write sources. These development, test or recovery copies are immediately available and have the same native Exadata performance and scale as the source databases, Oracle said.
Watch LIVE TV, Get Stock Market Updates, Top Business, IPO and Latest News on NDTV Profit.