The Most Popular Databases in 2026: A Live Popularity Ranking
Picking a database is a ten-year bet: it decides who you can hire, what your tools support, and how easy help is to find at 3 a.m. This index tracks where real demand is heading for 64 systems, re-ranked every month, so you see the trend before you commit.
| #▲ | Database ▲ | Score i ▼ | 12-mo interest i ▼ | Trend i |
|---|---|---|---|---|
| 1 |  PostgreSQLRelational | ▲6% | ||
PostgreSQLOpen-source object-relational database system known for standards compliance and extensibility. Sub-Scores | ||||
| 2 |  DatabricksAnalytical | ▲8% | ||
DatabricksLakehouse platform unifying data warehousing and AI on open table formats. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 3 |  RedisDoc/KV | ▼19% | ||
RedisIn-memory key-value data store used as a cache, message broker and database. Sub-Scores | ||||
| 4 |  SnowflakeAnalytical | ▼2% | ||
SnowflakeCloud data platform with separated storage and compute for analytics workloads. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 5 |  MySQLRelational | ▼16% | ||
MySQLWidely deployed open-source relational database developed by Oracle since 2010. Sub-Scores | ||||
| 6 |  Microsoft SQL ServerRelational | ▼34% | ||
Microsoft SQL ServerMicrosoft's commercial relational database for OLTP, BI and analytics workloads. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 7 |  MongoDBDoc/KV | ▼26% | ||
MongoDBDocument database storing JSON-like BSON documents with a distributed architecture. Sub-Scores | ||||
| 8 |  Oracle DatabaseRelational | ▼14% | ||
Oracle DatabaseOracle's flagship commercial multi-model relational database (we track the database product itself, separate from Oracle Corp overall). Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 9 |  ElasticsearchSearch | ▼3% | ||
ElasticsearchDistributed search and analytics engine built on Apache Lucene (tracked separately from OpenSearch). Sub-Scores | ||||
| 10 |  Google BigQueryAnalytical | ▼9% | ||
Google BigQueryGoogle Cloud's serverless data warehouse. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 11 |  SQLiteRelational | ▲21% | ||
SQLiteEmbedded serverless SQL database engine shipped inside applications on every platform. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 12 |  PrometheusTime-series | ▼25% | ||
PrometheusOpen-source monitoring system with a built-in time-series database, a CNCF graduated project. Sub-Scores | ||||
| 13 |  ClickHouseAnalytical | ▲21% | ||
ClickHouseOpen-source columnar OLAP database for real-time analytics. Sub-Scores | ||||
| 14 |  SupabaseprovisionalNewSQL/Edge | · | ||
SupabaseOpen-source Firebase alternative: hosted Postgres with auth, storage, and realtime APIs. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 15 |  IBM Db2Heritage | ▼24% | ||
IBM Db2IBM's commercial relational database family for enterprise and mainframe workloads. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 16 |  DuckDBAnalytical | — | ||
DuckDBIn-process analytical SQL database, often described as SQLite for analytics. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 17 |  MariaDBRelational | ▼4% | ||
MariaDBCommunity-developed fork of MySQL with independent governance and storage engines. Sub-Scores | ||||
| 18 |  Neo4jGraph/WC | ▼4% | ||
Neo4jNative property-graph database with the Cypher query language. Sub-Scores | ||||
| 19 |  Amazon DynamoDBDoc/KV | ▼33% | ||
Amazon DynamoDBAWS's fully managed serverless key-value and document database. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 20 |  Firebase (Firestore & Realtime Database)Doc/KV | ▼43% | ||
Firebase (Firestore & Realtime Database)Google's mobile/web backend databases: Cloud Firestore and the Firebase Realtime Database. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 21 |  TeradataHeritage | ▲19% | ||
TeradataCommercial MPP relational platform for large-scale enterprise data warehousing. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 22 |  Amazon RedshiftAnalytical | ▼47% | ||
Amazon RedshiftAWS's managed columnar data warehouse. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 23 |  WeaviateVector | · | ||
WeaviateOpen-source AI-native vector database with hybrid search. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 24 |  QdrantVector | · | ||
QdrantOpen-source vector database and similarity search engine written in Rust. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 25 |  SplunkSearch | ▼37% | ||
SplunkCommercial platform for indexing, searching and analyzing machine data, owned by Cisco. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 26 |  PineconeVector | · | ||
PineconeProprietary managed vector database for similarity search. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 27 |  Azure Cosmos DBDoc/KV | ▼36% | ||
Azure Cosmos DBMicrosoft Azure's globally distributed multi-model database service. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 28 |  OpenSearchSearch | ▼6% | ||
OpenSearchAWS-initiated fork of Elasticsearch and Kibana, now governed by the Linux Foundation. Sub-Scores | ||||
| 29 |  SAP HANAHeritage | ▼18% | ||
SAP HANASAP's in-memory column-oriented relational database platform. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 30 |  TrinoAnalytical | ▼45% | ||
TrinoOpen-source distributed SQL query engine for federated analytics across data lakes and databases. Sub-Scores | ||||
| 31 |  ValkeyprovisionalDoc/KV | ▼44% | ||
ValkeyLinux Foundation fork of Redis created in 2024 after Redis's license change, backed by AWS and Google. Sub-Scores | ||||
| 32 |  Apache CassandraGraph/WC | ▼39% | ||
Apache CassandraApache wide-column store designed for high write throughput across commodity clusters. Sub-Scores | ||||
| 33 |  InfluxDBTime-series | ▼25% | ||
InfluxDBPurpose-built time-series database from InfluxData. Sub-Scores | ||||
| 34 |  CockroachDBNewSQL/Edge | ▼4% | ||
CockroachDBDistributed SQL database with PostgreSQL wire compatibility. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 35 |  MilvusVector | ▼34% | ||
MilvusOpen-source vector database built for embedding similarity search, an LF AI & Data project. Sub-Scores | ||||
| 36 |  AlgoliaSearch | ▼15% | ||
AlgoliaProprietary hosted search-as-a-service API. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 37 |  MemcachedDoc/KV | ▼22% | ||
MemcachedDistributed in-memory key-value cache for small chunks of arbitrary data. Sub-Scores | ||||
| 38 |  TimescaleDBTime-series | ▼32% | ||
TimescaleDBPostgreSQL extension that turns Postgres into a time-series database. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 39 |  VerticaHeritage | ▼18% | ||
VerticaCommercial columnar analytics database, now part of OpenText. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 40 |  ChromaprovisionalVector | · | ||
ChromaOpen-source embedding database aimed at LLM applications. Sub-Scores | ||||
| 41 |  Apache SolrSearch | ▼37% | ||
Apache SolrApache search platform built on Lucene. Sub-Scores | ||||
| 42 |  Azure SQL DatabaseRelational | ▼23% | ||
Azure SQL DatabaseMicrosoft's cloud-native SQL Server-compatible database service built on the Hyperscale backend. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 43 |  VictoriaMetricsTime-series | · | ||
VictoriaMetricsOpen-source time-series database and monitoring backend, Prometheus-compatible. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 44 |  Google Cloud SpannerNewSQL/Edge | ▼20% | ||
Google Cloud SpannerGoogle Cloud's globally distributed, strongly consistent relational database. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 45 |  SAP ASE (Sybase)Heritage | ▼26% | ||
SAP ASE (Sybase)SAP's enterprise relational database, formerly Sybase Adaptive Server Enterprise. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 46 |  kdb+Time-series | ▼17% | ||
kdb+KX's commercial columnar time-series database used heavily in finance. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 47 |  LanceDBprovisionalVector | · | ||
LanceDBEmbedded open-source vector database built on the Lance columnar format. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 48 |  IBM InformixHeritage | — | ||
IBM InformixIBM's embedded and enterprise relational database with time-series extensions. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 49 |  Apache HiveAnalytical | ▼36% | ||
Apache HiveApache data warehouse software providing SQL over large distributed datasets. Sub-Scores | ||||
| 50 |  Amazon AuroraRelational | ▼43% | ||
Amazon AuroraAWS-built relational engine with MySQL/PostgreSQL compatibility on a proprietary log-structured distributed storage layer. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 51 |  etcdDoc/KV | ▼27% | ||
etcdDistributed key-value store providing consistent configuration and coordination for clusters, core to Kubernetes. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 52 |  ArangoDBGraph/WC | ▼32% | ||
ArangoDBMulti-model database with graph as its primary model plus document and key-value access. Sub-Scores | ||||
| 53 |  TiDBNewSQL/Edge | ▼31% | ||
TiDBDistributed MySQL-compatible HTAP database from PingCAP. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 54 |  ScyllaDBGraph/WC | ▼31% | ||
ScyllaDBC++ rewrite of Cassandra targeting low latency and high throughput per node. Sub-Scores | ||||
| 55 |  Apache CouchDBDoc/KV | ▼27% | ||
Apache CouchDBApache document database with an HTTP API and multi-master replication. Sub-Scores | ||||
| 56 |  CouchbaseDoc/KV | ▼52% | ||
CouchbaseDistributed document database combining key-value access with SQL-like querying (SQL++). Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 57 |  Apache HBaseGraph/WC | ▼43% | ||
Apache HBaseApache wide-column store built on HDFS, modeled after Google Bigtable. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 58 |  FirebirdHeritage | ▼10% | ||
FirebirdOpen-source relational database descended from Borland InterBase. Sub-Scores | ||||
| 59 |  YugabyteDBNewSQL/Edge | ▼15% | ||
YugabyteDBDistributed SQL database with a PostgreSQL-compatible API. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 60 |  QuestDBTime-series | · | ||
QuestDBOpen-source time-series database with SQL and high-throughput ingestion. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 61 |  NeonprovisionalNewSQL/Edge | · | ||
NeonServerless PostgreSQL platform with storage/compute separation and database branching. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 62 |  Turso (libSQL)provisionalNewSQL/Edge | · | ||
Turso (libSQL)Edge-hosted database platform built on libSQL, an open-contribution fork of SQLite. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 63 |  SurrealDBprovisionalNewSQL/Edge | · | ||
SurrealDBMulti-model database combining document, graph and vector features with the SQL-like SurrealQL. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
| 64 |  H2 DatabaseHeritage | ▼12% | ||
H2 DatabaseOpen-source embedded and in-memory Java SQL database. Some signal families don't apply to this system, so its score re-balances across the families that do. Sub-Scores | ||||
Database Popularity Trends: 12-Month View i
These lines show public interest until the index has enough score history of its own. Compare up to 8 systems: click a name to hide its line, hit × to drop it.
Most Popular Databases by Category
One leaderboard flattens nine different jobs. A vector store and a mainframe-era relational engine are never competing for the same slot, so the category pills above the table split the ranking into the races that actually happen. Here is what each one covers.
What Each Category Ranks
- Relational. The classic SQL engines that still run most transactional systems. Popularity moves slowly here and rewards decades of tooling, drivers, and deep hiring pools.
- Analytical. Warehouses and lakehouses built for scans rather than single-row lookups. The fastest-moving enterprise category in the index.
- Vector. Embedding stores for search and retrieval. Young, crowded, and the one category where a single year of momentum can rewrite the order.
- Document and key-value. Schema-light stores and caches. Developer signals dominate, because adoption starts in a side project long before it reaches procurement.
- Time-series. Engines tuned for metrics and event streams. Observability budgets drive this group more than application teams do.
- Graph and wide-column. Specialist engines chosen for one shape of query. Small categories, so a single strong release shows up clearly.
- Search. Inverted-index engines that sit beside a primary database. Ranked separately because they are rarely the system of record.
- NewSQL and edge. Distributed SQL and embedded engines. Heavy developer interest, thinner hiring demand, and the widest gap between our two sub-scores.
- Heritage. Systems past their commercial peak that still run critical workloads. Low momentum, real installed base, and the hardest group to staff.
Frequently Asked Questions
Which database is the most popular right now?
PostgreSQL leads the Jul 2026 index at 117.9, ahead of Databricks (114.1) and Redis (114.1).
What feeds the score?
Six families of public demand signals: hiring activity, reader interest, infrastructure activity, developer adoption, community questions, and technical discussion. We publish the families and the full monthly results. The exact sources, weights, and formulas stay private.
Why isn't the #1 database scored 100?
100 marks the average database, and the leader sits above the average. If 100 were pinned to the leader instead, every other score would drop whenever #1 surged, even with nothing else changing. Anchoring to the average avoids that.
How often does this update?
Monthly, in the first days of the month. Every chart carries the data month it was computed from.
Is Databricks really a database?
Databricks is a lakehouse platform with a database engine at its core, and database rankings have long listed it alongside warehouses like Snowflake and BigQuery. Each row carries a category tag, so filter by Relational if you want the classic engines only.
How We Rank Database Popularity
Every month we collect several dozen measurements per database from public sources and roll them into six weighted signal families. We keep the exact sources, weights, and formulas private. That is what makes the index hard to game. Two sub-scores read the same data from different angles: developer activity covers code, adoption, and discussion; market demand covers hiring and mainstream interest.
What This Ranking Does Not Measure
- Installed base. Nothing public counts production servers. This index measures attention and demand, which lead adoption.
- Question volume. Community Q&A partly rewards the databases people struggle with, so it carries one of the smallest weights in the blend.
- Job ads lag reality by months, and enterprise systems are louder there than open-source ones. The two sub-scores show each world separately.
- Ecosystems blur at the edges. Wire-compatible services ride their parent database’s drivers, and no ranking can split that traffic; ours re-balances the compatibles across the signals they do own. Download registries also skew open-source, so the hiring and readership signals carry the enterprise side that registries miss.
- Adjacent ranks are noise. Treat #7 vs #8 as a tie. Treat #7 vs #15 as a real gap.
Red9 is a database consulting firm.
If the databases behind your product deserve senior eyes, see our consulting and managed services.
Talk to a Senior DBA ➜