Oracle Sharding distributes segments of a data set across many databases (shards) on different computers, on-premises or in cloud. It enables globally distributed, linearly scalable, multimodel databases. It requires no specialized hardware or software. Oracle Sharding does all this while rendering the strong consistency, full power of SQL, support for structured and unstructured data, and the Oracle Database ecosystem. It meets data sovereignty requirements and support applications that require low latency and high availability.
See how Oracle enables hyperscalability across globally distributed databases.
Oracle BlueKai Data Management Platform uses Oracle Sharding to store petabytes of data to scale to millions of transactions per second with a single database.
Eliminate performance bottlenecks while enabling linearly scaled database performance.How Oracle builds internet scale applications with Oracle Database (43:48)
Ensures that data stays in a given geographical location. Sharding facilitates a single global database, with data distributed across multiple regions.Learn how to build a global database
Use a shared-nothing architecture to meet data sovereignty and data proximity requirements and to protect databases from unplanned downtime.Multicloud reference architecture
All shards are presented to an application as a single logical database, speeding query response time on extremely large datasets.Learn more
Data can be loaded on all shards in parallel at extremely high rates, which powers linearly scalable architecture for data ingest. Additionally, Oracle Sharding can be used for Event Stream Processing, Internet of Things, Log Analytics, metric store, and time series data, eliminating the need for a separate data pipeline.Loading data into a sharded database
Oracle Database is a converged database. It can store relational data and other types of unstructured or semistructured data, like Text, JSON, Graph, and Spatial. Oracle Sharding can be used to store massive amounts of structured and unstructured data and to eliminate data fragmentation.Sharding JSON documents collection
Oracle Sharding automatically places data on the desired shard, saving time and eliminating manual data preparation.
Shards can be added and removed, and data can be resharded without any downtime or data loss.
A shard catalog allows customers to manage all shards in a sharded database as a single, logical database.
Oracle Sharding is natively supported by Oracle Database client drivers. Developers can treat a sharded database as a single logical database, simplifying application development.
A sharded database has protection from unplanned outages and provides mechanisms for online planned operations, enabling applications to meet stringent availability SLAs.
Oracle Database 21c allows storing of all data and redo in local persistent memory (PMem). SQL runs directly on data stored in a mapped persistent memory file system, eliminating the IO code path to traditional disk storage.
Oracle Sharding eliminates complex data preparation for internet-facing consumer applications, financial and mobile payment applications, and large-scale SaaS applications.
Oracle Sharding provides linear scalability of transactions, better application SLAs, and strict data consistency.
Oracle Sharding speeds applications that require massively parallel processing for large data volumes across many compute instances.
Oracle Sharding powers data ingest at extreme speeds for Internet of Things (IoT) applications.
Oracle Sharding supports applications which require very high SLAs often met with fault isolation.
Oracle Sharding allows you to deploy a database that supports extreme scale-out with complete data isolation distributed across a multicloud environment.