Vercel KV

Fully managed multi-model NoSQL service with wide-column, time-series, and search data models, petabyte-scale storage per table, secondary indexes, reserved and pay-per-request billing modes, and single-digit-millisecond latency

Fully managed JSON document database with MongoDB 3.6, 4.0, and 5.0 wire compatibility, separation of compute and storage up to 128 TiB, and automatic six-way replication across three Availability Zones

Fast, flexible key-value and document database with single-digit-millisecond latency at any scale, on-demand and provisioned capacity, and global tables

Fully managed in-memory data store and cache compatible with Redis and Memcached, used for session stores, leaderboards, and sub-millisecond read latency

Durable in-memory database compatible with Redis OSS and Valkey that delivers microsecond reads, single-digit millisecond writes, and Multi-AZ transaction log durability for both caching and primary database workloads

Globally distributed multi-model NoSQL database with SQL, MongoDB, Cassandra, Gremlin, and Table APIs, turnkey geo-replication, and five tunable consistency levels

Fully managed in-memory data store powered by Redis with Basic, Standard, Premium, and Enterprise tiers, geo-replication, Redis Modules, and clustering for horizontal scaling

Fully-managed NoSQL document database built on open-source DocumentDB tech and designed for hybrid + multicloud flexibility — MongoDB driver compatible with native vector search and embeddings, distinct from Cosmos DB by being open-source-rooted with portable APIs rather than the proprietary Cosmos SDK

Globally-replicated eventually-consistent key-value store accessible from Workers with low-latency edge reads, 25 MiB value limit, per-key TTL, metadata, and bulk write APIs — the native KV primitive for caching, session storage, and feature flags at Cloudflare's edge

Serverless NoSQL document database with real-time listeners, offline support for mobile and web SDKs, ACID transactions, and automatic multi-region replication

High-throughput wide-column NoSQL database that powers Google Search, Maps, and Gmail internally — single-digit millisecond reads at petabyte scale, HBase-compatible API, autoscaling node clusters

Managed in-memory data store compatible with Redis and Memcached, with Basic and Standard tiers, automatic failover, in-transit and at-rest encryption, and AUTH-protected access

Managed Redis-compatible key-value store — pub/sub, caching, sessions, queue backing for sidekiq / celery / bull workers

Managed Redis and Memcached service with master-standby, cluster, and read-replica deployment modes, online scaling, automatic backup to OBS, and VPC-isolated access across Huawei Cloud

Managed JSON document database based on Apache CouchDB with HTTP/JSON API, full-text search, offline sync, geospatial queries, and automatic cross-region replication

Managed key-value and document NoSQL database with ACID transactions, flexible schema, global active-active replication, on-demand and provisioned read/write capacity, and single-digit millisecond latency at any scale

Managed in-memory cache service announced at Oracle CloudWorld 2025 — Redis-compatible cluster management with multi-AZ replication, automated patching, and TLS-by-default, sitting alongside Autonomous Database for hot-path acceleration of high-throughput read workloads

Managed in-memory key-value store with Redis and Memcached engines for caching and session-store workloads, with HA replication and Germany-only data residency

Managed MongoDB document database with replica-set HA, automated backups, and German-only data residency

Managed in-memory key-value store with HA replication and persistence options for caching and session-store workloads — Germany-only data residency