Enterprise Platform
Distributed Caching
Move hot data and read-heavy workloads to a distributed in-memory tier that serves applications with sub-millisecond latency and high throughput, without application code changes for query caching.
Speed without re-architecting
The distributed in-memory cache layer sits between your apps and the core transaction processing system. It moves hot data into a shared-nothing, horizontally scalable memory tier, delivering sub-millisecond reads with ANSI SQL and ACID guarantees.
Benefits
Why distributed caching?
Ultra-low latency at scale
Sub-millisecond data access across a distributed cluster that scales linearly to millions of transactions per second and petabytes of data.
High availability to run 24/7
Stay open when a site fails. Geo-distributed data center replication runs active-active or active-passive, with rack awareness and configurable backup copies.
Offload from the core
Absorb read-heavy and repetitive traffic in front of the mainframe or core database, cutting MIPS and licensing cost without changing the system of record.
Native MariaDB integration
Zero code changes when deployed as a query result cache in front of your MariaDB database. Works with any system of record: Oracle, SQL Server or mainframe.
Capabilities
Built for live data caching
Horizontal scalability
Scale linearly through partitioning data across the cluster and auto rebalancing when new nodes are added to the cluster.
Strong consistency
Read-through, write-through and write-behind keep the key-value cache consistent with your system of record.
Multi-model with ANSI SQL
One cluster holds key-value, relational, JSON and vector data. Query key-value and relational data with ANSI SQL, so existing tools and skills carry over.
Enterprise-ready operations
Your Ops team runs one console. Deploy on Kubernetes, monitor cache clusters from MariaDB Enterprise Manager, and stay available with active-active replication.
Also included:
Native disk persistence
Key-value and SQL APIs
Semantic cache
Agent memory
Point-in-time recovery
Kubernetes Operator deployment
OpenTelemetry observability
Encryption in-transit/at-rest
RBAC and audit logging
Zero-downtime rolling upgrades
Looking for Enterprise Platform pricing?
Recommended resources
Get started
Start building with MariaDB
One real-time database platform you can trust, for everything you run today and everything you build next.
Frequently asked questions
The MariaDB distributed cache is an in-memory cache and compute layer based on MariaDB GridGain. It sits between your applications and systems of record, serving hot data in sub-milliseconds.
The in-memory architecture partitions data across the RAM of many servers so reads and computations run in parallel. Adding nodes rebalances data automatically, linearly scaling throughput and capacity.
Both cache data in memory. Redis is a single-threaded key-value datastore, while MariaDB's distributed cache is multi-threaded and adds ANSI SQL, ACID transactions and disk-based persistence.
No. Used as a query cache with the MaxScale intelligent proxy, it acts as a transparent layer in front of the MariaDB database.
Yes. MariaDB Enterprise Manager auto-discovers the cache clusters alongside MariaDB Server and MaxScale, with single sign-on into GridGain Control Center for deeper cluster-specific management.