I love Datalog, why hasn't everyone used this already?

Datalevin is a simple durable Datalog database. Here's what a Datalog query looks like in Datalevin:

(d/q '[:find  ?name ?total
       :in    $ ?year
       :where [?sales :sales/year ?year]
              [?sales :sales/total ?total]
              [?sales :sales/customer ?customer]
              [?customer :customers/name ?name]]
      (d/db conn) 2024)

:question: Why

The rationale is to have a simple, fast and open source Datalog query engine running on durable storage.

It is our observation that many developers prefer the flavor of Datalog popularized by Datomic® over any flavor of SQL, once they get to use it. Perhaps it is because Datalog is more declarative and composable than SQL, e.g. the automatic implicit joins seem to be its killer feature. In addition, the recursive rules feature of Datalog makes it suitable for graph processing and deductive reasoning.

The feature set of Datomic® may not be a good fit for some use cases. One thing that may confuse some users is its temporal features. To keep things simple and familiar, Datalevin behaves the same way as most other databases: when data are deleted, they are gone. Datalevin also follows the widely accepted principles of ACID, instead of introducing unusual semantics.

Datalevin started out as a port of Datascript in-memory Datalog database to LMDB for persistence. We then added a cost-based query optimizer, resulting in query performance competitive against SQL RDBMS such as PostgreSQL.

Datalevin can be used as a library, embedded in applications to manage state, e.g. used like SQLite; or it can run in a networked client/server mode (default port is 8898) with full-fledged role-based access control (RBAC) on the server, e.g. used like PostgreSQL.

Datalevin relies on the robust ACID transactional database features of LMDB. Designed for concurrent read intensive workloads, LMDB also performs well in writing large values (> 2KB). Therefore, it is fine to store documents in Datalevin.

Datalevin can be used as a fast key-value store for EDN data. The native EDN data capability of Datalevin should be beneficial for Clojure programs.

Moreover, Datalevin has a built-in full-text search engine that has competitive search performance.

Additional materials:

• Competing for the JOB with a Triplestore
• If I had to Pick One: Datalevin
• T-Wand: Beat Lucene in Less Than 600 Lines of Code
• Writing C code in Java/Clojure: GraalVM specific programming
• 2020 London Clojurians Meetup

:truck: Installation

As a Clojure library, Datalevin is simple to add as a dependency to your Clojure project. There are also several other options. Please see details in Installation Documentation

:birthday: Upgrade

Please read Upgrade Documentation for information regarding upgrading your existing Datalevin database from older versions.

:tada: Usage

Datalevin is aimed to be a versatile database.

Use as a Datalog store

In addition to our API doc, since Datalevin has almost the same Datalog API as Datascript, which in turn has almost the same API as Datomic®, please consult the abundant tutorials, guides and learning sites available online to learn about the usage of Datomic® flavor of Datalog.

Here is a simple code example using Datalevin:

(require '[datalevin.core :as d])

;; Define an optional schema.
;; Note that pre-defined schema is optional, as Datalevin does schema-on-write.
;; However, attributes requiring special handling need to be defined in schema,
;; e.g. many cardinality, uniqueness constraint, reference type, and so on.
(def schema {:aka  {:db/cardinality :db.cardinality/many}
             ;; :db/valueType is optional, if unspecified, the attribute will be
             ;; treated as EDN blobs, and may not be optimal for range queries
             :name {:db/valueType :db.type/string
                    :db/unique    :db.unique/identity}})

;; Create DB on disk and connect to it, assume write permission to create given dir
(def conn (d/get-conn "/tmp/datalevin/mydb" schema))
;; or if you have a Datalevin server running on myhost with default port 8898
;; (def conn (d/get-conn "dtlv://myname:mypasswd@myhost/mydb" schema))

;; Transact some data
;; Notice that :nation is not defined in schema, so it will be treated as an EDN blob
(d/transact! conn
             [{:name "Frege", :db/id -1, :nation "France", :aka ["foo" "fred"]}
              {:name "Peirce", :db/id -2, :nation "france"}
              {:name "De Morgan", :db/id -3, :nation "English"}])

;; Query the data
(d/q '[:find ?nation
       :in $ ?alias
       :where
       [?e :aka ?alias]
       [?e :nation ?nation]]
     (d/db conn)
     "fred")
;; => #{["France"]}

;; Retract the name attribute of an entity
(d/transact! conn [[:db/retract 1 :name "Frege"]])

;; Pull the entity, now the name is gone
(d/q '[:find (pull ?e [*])
       :in $ ?alias
       :where
       [?e :aka ?alias]]
     (d/db conn)
     "fred")
;; => ([{:db/id 1, :aka ["foo" "fred"], :nation "France"}])

;; Close DB connection
(d/close conn)

Use as a key-value store

Datalevin packages the underlying LMDB database as a convenient key-value store for EDN data.

(require '[datalevin.core :as d])
(import '[java.util Date])

;; Open a key value DB on disk and get the DB handle
(def db (d/open-kv "/tmp/datalevin/mykvdb"))
;; or if you have a Datalevin server running on myhost with default port 8898
;; (def db (d/open-kv "dtlv://myname:mypasswd@myhost/mykvdb" schema))

;; Define some table (called "dbi", or sub-databases in LMDB) names
(def misc-table "misc-test-table")
(def date-table "date-test-table")

;; Open the tables
(d/open-dbi db misc-table)
(d/open-dbi db date-table)

;; Transact some data, a transaction can put data into multiple tables
;; Optionally, data type can be specified to help with range query
(d/transact-kv
  db
  [[:put misc-table :datalevin "Hello, world!"]
   [:put misc-table 42 {:saying "So Long, and thanks for all the fish"
                        :source "The Hitchhiker's Guide to the Galaxy"}]
   [:put date-table #inst "1991-12-25" "USSR broke apart" :instant]
   [:put date-table #inst "1989-11-09" "The fall of the Berlin Wall" :instant]])

;; Get the value with the key
(d/get-value db misc-table :datalevin)
;; => "Hello, world!"
(d/get-value db misc-table 42)
;; => {:saying "So Long, and thanks for all the fish",
;;     :source "The Hitchhiker's Guide to the Galaxy"}


;; Range query, from unix epoch time to now
(d/get-range db date-table [:closed (Date. 0) (Date.)] :instant)
;; => [[#inst "1989-11-09T00:00:00.000-00:00" "The fall of the Berlin Wall"]
;;     [#inst "1991-12-25T00:00:00.000-00:00" "USSR broke apart"]]

;; This returns a PersistentVector - e.g. reads all data in JVM memory
(d/get-range db misc-table [:all])
;; => [[42 {:saying "So Long, and thanks for all the fish",
;;          :source "The Hitchhiker's Guide to the Galaxy"}]
;;     [:datalevin "Hello, world!"]]

;; This allows you to iterate over all DB keys inside a transaction.
;; You can perform writes inside the transaction.
;; kv is of of type https://www.javadoc.io/doc/org.lmdbjava/lmdbjava/latest/org/lmdbjava/CursorIterable.KeyVal.html
;; Avoid long-lived transactions. Read transactions prevent reuse of pages freed by newer write transactions, thus the database can grow quickly.
;; Write transactions prevent other write transactions, since writes are serialized.
;; LMDB advice: http://www.lmdb.tech/doc/index.html
;; Conclusion: It's ok to have long transactions if using a single thread.
(d/visit db misc-table
            (fn [kv]
               (let [k (d/read-buffer (d/k kv) :data)]
                  (when (= k 42)
                    (d/transact-kv db [[:put misc-table 42 "Don't panic"]]))))
              [:all])

(d/get-range db misc-table [:all])
;; => [[42 "Don't panic"] [:datalevin "Hello, world!"]]

;; Delete some data
(d/transact-kv db [[:del misc-table 42]])

;; Now it's gone
(d/get-value db misc-table 42)
;; => nil

;; Close key value db
(d/close-kv db)

:green_book: Documentation

Please refer to the API documentation for more details. You may also consult online materials for Datascript or Datomic®, as the Datalog API is similar.

:rocket: Status

Datalevin is extensively tested with property-based testing. It is also used in production at Juji.

Running the benchmark suite adopted from Datascript, which write 100K random datoms in several conditions, and run several queries on them, on a Ubuntu Linux server with an Intel i7 3.6GHz CPU and a 1TB SSD drive, here is how it looks.

In all benchmarked queries, Datalevin is the fastest among the three tested systems, as Datalevin has a cost based query optimizer while Datascript and Datomic do not. Datalevin also has a caching layer for index access.

See here for a detailed analysis of the results.

We also compared Datalevin and PostgreSQL in handling complex queries, using Join Order Benchmark. On a MacBook Pro, Apple M3 chip with 12 cores, 30 GB memory and 1TB SSD drive, here's the average times:

Datalevin is about 1.3X faster than PostgreSQL on average in running the complex queries in this benchmark. The gain is mainly in better query execution time due to higher quality of generated query plans.

For more details, see here.

:earth_americas: Roadmap

These are the tentative goals that we try to reach as soon as we can. We may adjust the priorities based on feedback.

• 0.4.0 Native image and native command line tool. [Done 2021/02/27]
• 0.5.0 Native networked server mode with role based access control. [Done 2021/09/06]
• 0.6.0 As a search engine: full-text search across database. [Done 2022/03/10]
• 0.7.0 Explicit transactions, lazy results loading, and results spill to disk when memory is low. [Done 2022/12/15]
• 0.8.0 Long ids; composite tuples; enhanced search engine ingestion speed. [Done 2023/01/19]
• 0.9.0 New Datalog query engine with improved performance. [Done 2024/03/09]
• 1.0.0 New rule evaluation algorithm and incremental view maintenance.
• 1.1.0 Option to store data in compressed form.
• 1.2.0 Extensible de/serialization for arbitrary data.
• 2.0.0 Vector indexing and similarity search.
• 2.1.0 Automatic document indexing.
• 2.2.0 Extended full-text search syntax.
• 3.0.0 JSON API and library/client for popular languages.
• 4.0.0 Transaction log storage and access API.
• 4.1.0 Read-only replicas for server.
• 5.0.0 Distributed mode.

:floppy_disk: Differences from Datascript

Datascript is developed by Nikita Prokopov that "is built totally from scratch and is not related by any means to" Datomic®. Datalevin started out as a port of Datascript to LMDB, but differs from Datascript in more significant ways than just the difference in data durability and running mode:

• Datalevin has a cost-based query optimizer, so queries are truly declarative and clause ordering does not affect query performance.

• Datalevin is not an immutable database, and there is no "database as a value" feature. Since history is not kept, transaction ids are not stored.

• Datoms in a transaction are committed together as a batch, rather than being saved by with-datom one at a time.

• ACID transaction and rollback are supported.

• Lazy results set and spill to disk are supported.

• Entity and transaction integer ids are 64 bits long, instead of 32 bits.

• Respects :db/valueType. Currently, most Datomic® value types are supported, except uri. Values of the attributes that are not defined in the schema or have unspecified types are treated as EDN blobs, and are de/serialized with nippy.

• In addition to composite tuples, Datalevin also supports heterogeneous and homogeneous tuples.

• More query functions, such as like and not-like that are similar to LIKE and NOT LIKE operators in SQL; in and not-in that are similar to IN and NOT IN operators in SQL, among others.

• Has a value leading index (VAE) for datoms with :db.type/ref type attribute; The attribute and value leading index (AVE) is enabled for all datoms, so there is no need to specify :db/index, similar to Datomic® Cloud. Does not have AEV index, in order to save storage and improve write speed.

• Stored transaction functions of :db/fn should be defined with inter-fn, for function serialization requires special care in order to support GraalVM native image. It is the same for functions that need to be passed over the wire to server or babashka.

• Attributes are stored in indices as integer ids, thus attributes in index access are returned in attribute creation order, not in lexicographic order (i.e. do not expect :b to come after :a). This is the same as Datomic®.

• Has no features that are applicable only for in-memory DBs, such as DB as an immutable data structure, DB pretty print, etc.

:baby: Limitations

• Attribute names have a length limitation: an attribute name cannot be more than 511 bytes long, due to LMDB key size limit.

• Because keys are compared bitwise, for range queries to work as expected on an attribute, its :db/valueType should be specified.

• Floating point NaN cannot be stored.

• Big integers do not go beyond the range of [-2^1015, 2^1015-1], the unscaled value of big decimal has the same limit.

• The maximum individual value size is 2GB. Limited by the maximum size of off-heap byte buffer that can be allocated in JVM.

• The total data size of a Datalevin database has the same limit as LMDB's, e.g. 128TB on a modern 64-bit machine that implements 48-bit address spaces.

• Currently supports Clojure on JVM 8 or the above, but adding support for other Clojure-hosting runtime is possible, since bindings for LMDB exist in almost all major languages and available on most platforms.

:shopping: Alternatives

If you are interested in using the dialect of Datalog pioneered by Datomic®, here are your current options:

• If you need time travel and cloud features backed by the company that maintains Clojure, and there is no need to see the source code, you may try Datomic®.

• If you need mainly an in-memory store with optional durability, that has almost the same API as Datomic®, you may try Datascript.

• If you need a simple and versatile store with almost the same Datalog API as the above two and with a much greater query performance, you may try Datalevin, this project.

• If you need features such as bi-temporal models and SQL, you may try XTDB.

• If you need a graph database with an open world assumption, you may try Asami.

• If you need a durable store with some storage choices, you may try Datahike.

• There was also Eva, a distributed store, but it is no longer in active development.

:arrows_clockwise: Contact

We appreciate and welcome your contributions or suggestions. Please feel free to file issues or pull requests.

If commercial support is needed for Datalevin, talk to us.

You can talk to us in the #datalevin channel on Clojurians Slack.

License

Copyright © 2020-2024 Juji, Inc..

Licensed under Eclipse Public License (see LICENSE).