dbdb.io

dbdb.io (Database of Databases) lists database systems across 14 data models, spanning relational, key-value, document/XML, graph, object-oriented, vector, triplestore/RDF, column family/wide-column, array/matrix, network, multi-value, hierarchical, entity-relationship, and entity-attribute-value systems. Entries include both open-source and proprietary products, from long-established commercial databases to newly created ones. You can browse by data model or track systems through the site's leaderboards.

Coverage by data model

The site organizes its catalog around data models, and each model page shows how many systems it contains. According to the current listing:

Relational systems dominate the catalog by a wide margin, followed by key-value and document/XML. The presence of vector, triplestore/RDF, and array/matrix categories shows the catalog extends well beyond conventional transactional databases into analytical, semantic, and scientific workloads.

Open-source and proprietary systems

The catalog mixes licensing and governance models rather than restricting itself to one. The site maintains dedicated groupings such as Apache Foundation databases (systems controlled by the Apache Software Foundation) and Wisconsin Database Systems (from the University of Wisconsin), alongside individual entries that may be open source or proprietary.

The featured system illustrates the proprietary side: Amazon Redshift, described as a data warehouse product and part of Amazon Web Services, is labeled "Proprietary" and "OLAP," with a start year of 2012.

Established and emerging systems

Listings range from decades-old commercial products to recently created projects. The "Most Viewed" leaderboard over the last 30 days includes Oracle (a multi-model system tagged Key-Value, Graph, Triplestore/RDF, Document/XML, and Relational) and Derby (Relational), while the "Most Recent" leaderboard shows newer or newly modified entries such as OrioleDB, Turso, Xodus, voidDB, and CnosDB.

How the catalog is organized

Two structures make the catalog navigable:

• Browse by Data Model — groups every system under one of the 14 models above, so you can start from the category that matches your workload.
• The Leaderboards — updated daily, with views such as Most Recent (newly modified entries), Most Viewed (last 30 days), and Most Edited (last 30 days). The Most Viewed board also shows percentage change, indicating rising or falling interest.

A Featured System section, updated weekly, highlights one entry in more depth.

What this means for you

If you are looking for a specific class of database — say a vector store, a graph database, or a wide-column system — the data model index is the fastest entry point. If you want to see what is gaining attention or being actively maintained, the leaderboards surface recent activity and viewing trends. Because the catalog includes both open-source and proprietary systems, it works as a starting point for comparison, but you should confirm licensing, pricing, and deployment details on each system's own entry or vendor site.

dbdb.io is the Database of Databases, a curated reference guide to implementations of database systems, maintained by Carnegie Mellon University and described as "the encyclopedia of database systems." It is useful when you need a structured overview of a database product, want to compare systems by data model, or are looking for a starting point for research or tool selection. Entries are reference material rather than vendor documentation, so treat them as orientation, not as the final word on a product's current features.

What you can find on the site

The site organizes database systems into browsable entries and aggregate views:

• Individual entries for database systems, such as the featured entry on Amazon Redshift, which is described as a data warehouse product and part of Amazon Web Services, a relational DBMS designed specifically for that purpose, proprietary, OLAP, and in existence since 2012.
• Browse by data model, with counts of systems per model.
• Leaderboards, updated daily, tracking activity and interest.
• Thematic collections, such as databases named "Store," Wisconsin database systems, and databases controlled by the Apache Software Foundation.

Browse by data model

Data model is the primary organizing dimension. The site lists 14 models with the following counts:

Use this view when your constraint is the data model itself — for example, when you need a graph or vector store and want to see what exists in that category before narrowing by other criteria.

Leaderboards and what they measure

The leaderboards are updated daily and cover different signals, so read the label before drawing conclusions:

• Most Recent — newly modified entries, with a relative timestamp. Examples: OrioleDB (Relational, 3d ago), Turso (Relational, 3d ago), Xodus (Document / XML, 9d ago), voidDB (Key-Value, 9d ago), CnosDB (Relational, 10d ago).
• Most Viewed (last 30 days) — includes a percentage change, e.g. Oracle (Key-Value, Graph, Triplestore / RDF, Document / XML, Relational, −26%), Derby (Relational, −20%), Heroic (Key-Value, −48%), DynamoDB (Document / XML, Key-Value, −45%), TileDB (Array / Matrix, −1%).
• Most Edited (last 30 days) — edit counts, e.g. SereneDB (Relational, 2), CloverDB (Document / XML, 1), CnosDB (Relational, 1), ConceptBase.cc (Object-Oriented, 1), Dremio (Relational, 1).

"Most viewed" reflects attention, not quality or adoption; a falling percentage means fewer views than the prior period, not that a system is declining in use. "Most edited" reflects wiki activity, which often tracks how actively a community maintains its entry.

Who it is for

• Developers choosing or evaluating a database who want a model-first index and a quick factual summary.
• Students learning how database systems are classified and what categories exist.
• Researchers needing a citable overview of systems and their attributes.

How to use it in practice

1. Start from Browse by Data Model if your requirement is architectural, or from an entry directly if you already have a product in mind.
2. Open the entry and check the attributes it records — for example, Redshift's entry notes vendor, category, licensing, and year.
3. Cross-check anything decision-critical against the vendor's own documentation, since entries are reference summaries and may lag a product's current state.
4. Use the leaderboards to discover systems you did not know about, then evaluate them on their own entries.

Limitations to keep in mind

The site is an encyclopedia, not a benchmark or a buyer's guide. It does not publish pricing, performance results, or support commitments, and the counts and rankings reflect catalog contents and site activity rather than market share. For a purchase or migration decision, use dbdb.io to build your shortlist, then verify capabilities, licensing, and cost with the vendor.

Go to dbdb.io and use the Browse by Data Model section on the homepage. It lists 14 data models, each with a count of catalogued systems, and clicking a model takes you to the matching list of database entries. This is the fastest route when you already know the technical category you want — for example, a relational engine for an OLTP workload or a vector store for embeddings — rather than searching by product name.

What the data model categories contain

The homepage groups every catalogued system under one of 14 models. The counts shown are the number of systems currently filed under each model:

A system can appear under more than one model. Oracle, for instance, is tagged Key-Value, Graph, Triplestore / RDF, Document / XML, and Relational, so browsing a single model will not always give you a clean one-to-one mapping between a product and a category.

Steps to browse by model

1. Open dbdb.io. The Browse by Data Model block sits on the homepage alongside Featured System and The Leaderboards.
2. Pick the model that matches your workload. Relational, Key-Value, and Document / XML cover the largest share of the catalogue; Vector, Triplestore / RDF, and Column Family / Wide-Column are smaller but more specific.
3. Click the model name to open its list of systems.
4. Open an individual entry to read the description, model tags, and metadata for that system.

Expected result: you land on a filtered list of systems for that model, then drill into individual entries instead of scanning the whole catalogue.

Choosing which model to start from

• Structured transactions and SQL → Relational (473 systems), the largest category by a wide margin.
• Caching, sessions, simple lookups → Key-Value (248).
• Semi-structured or schema-flexible records → Document / XML (217).
• Relationship-heavy queries → Graph (91).
• Embeddings and similarity search → Vector (50).
• RDF and linked data → Triplestore / RDF (28).
• Analytical columnar workloads → Column Family / Wide-Column (28).

If your workload spans categories, check whether a candidate system carries multiple model tags before treating it as a single-model fit.

Other ways to move around the site

The homepage also offers:

• Featured System — a highlighted entry, updated weekly. At the time of writing it is Redshift, an OLAP relational data warehouse from Amazon, listed as proprietary and in use since 2012.
• The Leaderboards — updated daily, split into Most Recent (newly modified entries such as OrioleDB, Turso, Xodus, voidDB, CnosDB), Most Viewed over the last 30 days (Oracle, Derby, Heroic, DynamoDB, TileDB), and Most Edited over the last 30 days (SereneDB, CloverDB, CnosDB, ConceptBase.cc, Dremio).
• Curated groupings — for example Apache Foundation databases, databases named "Store", and university database systems such as the Wisconsin Database Systems collection.

Use the leaderboards when you want to see what is active or popular; use Browse by Data Model when you have already decided on a technical category.

Common snags

• Counts shift. The numbers reflect the catalogue at a given moment and change as entries are added or retagged, so treat them as scale indicators rather than fixed totals.
• Overlapping tags. Multi-model systems appear in several lists, which can make a category look larger or more crowded than the number of distinct products.
• Model is not the same as workload fit. A system filed under Relational may still be an OLAP warehouse rather than an OLTP engine, so read the entry before shortlisting.

The leaderboards on dbdb.io are activity rankings for the database entries in the encyclopedia, updated daily. They answer a different question than the main catalog: not "what database systems exist," but "which entries people are adding, reading, and revising right now." The page lists three boards — Most Recent, Most Viewed, and Most Edited — and each measures a distinct kind of activity over a stated time window.

The three leaderboards

Each board also carries its own refresh cadence: the leaderboards as a whole are marked "Updated Daily," while the Featured System section is marked "Updated Weekly."

Reading each board

Most Recent

This board lists entries by how recently they were modified, with a relative timestamp. In the snapshot shown, OrioleDB and Turso (both Relational) were modified 3 days ago, Xodus (Document / XML) 9 days ago, voidDB (Key-Value) 9 days ago, and CnosDB (Relational) 10 days ago. Use it to see what the community is actively maintaining rather than what is merely popular.

Most Viewed

This board covers the last 30 days and pairs each entry with a percentage change. The percentages are the part most likely to be misread: a negative figure is a decline in views, not a rank or a rating. In the snapshot, Oracle (tagged Key-Value, Graph, Triplestore / RDF, Document / XML, and Relational) was down 26%, Derby (Relational) down 20%, Heroic (Key-Value) down 48%, DynamoDB (Document / XML, Key-Value) down 45%, and TileDB (Array / Matrix) down 1%. Because every entry in that snapshot shows a decline, the percentages are best read as relative momentum within the board, not as absolute traffic levels.

Most Edited

This board counts edits over the last 30 days. The snapshot shows SereneDB (Relational) with 2 edits, and CloverDB (Document / XML), CnosDB (Relational), ConceptBase.cc (Object-Oriented), and Dremio (Relational) with 1 edit each. Low counts are normal here — this reflects how much correction or expansion a given entry needed, not how important the system is.

What the leaderboards are useful for

• Finding actively maintained entries. Most Recent surfaces systems whose pages were just updated, which is a reasonable starting point if you want current information.
• Spotting attention shifts. Most Viewed with its 30-day percentage change shows which entries are gaining or losing reader interest.
• Judging entry maturity. A high edit count can mean an entry is being actively fleshed out or corrected; a low one can mean it is stable or simply neglected.
• Cross-checking tags. Entries on these boards carry data-model tags (Relational, Key-Value, Document / XML, and so on), so a leaderboard row doubles as a quick pointer into the model-based browsing on the site.

Limits to keep in mind

The boards measure activity on dbdb.io itself, not the real-world popularity, performance, or adoption of a database system. A system with few page views may simply be well documented and rarely needing a visit. The page does not state how views are counted, whether unique visitors are distinguished from repeat visits, or what the baseline period for the percentage change is, so treat the numbers as directional signals rather than precise metrics. The Most Recent board also gives no fixed window — only relative timestamps — so its ordering shifts continuously as entries are touched.

The Featured System is a rotating spotlight on the dbdb.io homepage that highlights one database entry at a time. It shows a short description of the system along with its data model, license type, and the year it was introduced, and it is updated weekly. The current example is Amazon Redshift, described as a relational data warehouse product from Amazon Web Services, labeled Relational, Proprietary, and "Since 2012."

What the Featured System shows

The panel is a compact summary rather than a full article. Based on the current Redshift entry, it includes:

• A short description — what the system is and who makes it. Redshift is described as a data warehouse product developed by Amazon, part of its cloud platform, and a relational database management system designed specifically for that purpose.
• Data model — Redshift is tagged Relational.
• License — Redshift is tagged Proprietary.
• Age/origin — Redshift is tagged Since 2012.
• A "Read entry" link — the path to the full dbdb.io entry for that system.

The panel also carries the label Updated Weekly, which tells you the spotlight changes on a weekly cadence rather than staying fixed.

How it differs from the leaderboards

The Featured System and the leaderboards sit in the same area of the homepage but serve different purposes:

So the Featured System is editorial and descriptive — a single pick with context — while the leaderboards are ranked lists driven by site activity.

Why it is useful

Because dbdb.io catalogs hundreds of systems across 14 data models (Relational alone has 473 entries), browsing everything is impractical if you just want to see what exists. The Featured System gives you one concrete, described example per week, which is a low-effort way to encounter systems you might not search for directly. In the Redshift case, it surfaces a well-known cloud warehouse alongside its classification, so you immediately know where it sits in the taxonomy before deciding whether to open the full entry.

How to use it

1. Open the dbdb.io homepage.
2. Find the Featured System panel (it sits near the data-model browser and the leaderboards).
3. Read the summary to judge whether the system is relevant to you.
4. Select Read entry to open the full dbdb.io page for that system.
5. Check back weekly, since the pick rotates on that schedule.

If the featured system is not what you need, use Browse by Data Model instead — for example, Relational (473), Key-Value (248), or Document / XML (217) — or scan the leaderboards for recently modified or most-viewed entries.