Crux is an open source document database with bitemporal graph queries. Java, Clojure and HTTP APIs are provided.

Crux follows an unbundled architectural approach, which means that it is assembled from highly decoupled components through the use of semi-immutable logs at the core of its design. Logs can currently be stored in LMDB or RocksDB for standalone single-node deployments, or using Kafka for clustered deployments. Indexes can currently be stored using LMDB or RocksDB.

Crux is built for efficient bitemporal indexing of schemaless documents, and this simplicity enables broad possibilities for creating layered extensions on top, such as to add additional transaction, query, and schema capabilities. Crux does not currently support SQL but it does provide an EDN-based Datalog query interface that can be used to express a comprehensive range of SQL-like join operations as well as recursive graph traversals.

Crux has been available as a Public Alpha since 19^th April 2019. The Public Alpha period will continue until Crux is released as a Generally Available open source software product by JUXT later in 2019.

CircleCI Build

Latest Release

Unbundled Architecture

Crux embraces the transaction log as the central point of coordination when running as a distributed system. Use of a separate document log enables simple eviction of active and historical data to assist with technical compliance for information privacy regulations.

What do we have to gain from turning the database inside out? Simpler code, better scalability, better robustness, lower latency, and more flexibility for doing interesting things with data.

— Martin Kleppmann (2014)

This design makes it feasible and desirable to embed Crux nodes directly in your application processes, which reduces deployment complexity and eliminates round-trip overheads when running complex application queries.

Data Model

Crux is fundamentally a store of versioned EDN documents. The only requirement is that you specify a valid :crux.db/id key which links the documents to their corresponding entities. The fields within these documents are automatically indexed as Entity-Attribute-Value triples to support efficient graph queries. Document versions are indexed by valid-time (in addition to transaction-time) which allows you to model updates into the past, present or future.

Crux supports a Datalog query interface for reading data and traversing relationships across all documents. Queries are executed so that the results are lazily streamed from the underlying indexes. Queries can be made against consistent point-in-time snapshots of your database from any Crux node connected to the same transaction log, by specifying transaction-time and/or valid-time.

Documentation

Please visit our official documentation to get started with Crux.

Community & Contact

3rd Party Modules

• avisi-apps/crux-xodus - pure-JVM alternative to crux-rocksdb and crux-lmdb
• avisi-apps/crux-active-objects
  • TxLog implementation backed by Active Objects for use inside Atlassian Addons

Companies using Crux

• Oscaro (eCommerce)
• Avisi (AtlasCRM) - "Crux: Our Final Database Migration"
• Gnurdle (consulting)
• Yours? Let us know :)

crux@juxt.pro

Repo Layout

Crux is split across multiple projects which are maintained within this repository. crux-core contains the main functional components of Crux along with interfaces for the pluggable storage components (Kafka, LMDB, RocksDB etc.). Implementations of these storage options are located in their own projects.

Project directories are published to Clojars independently so that you can maintain granular dependencies on precisely the individual components needed for your application. Alternatively you can depend on crux-uberjar whilst in development to spend less time worrying about which parts of Crux you need now or in the future.

Core

crux-core

Cluster Node storage

crux-kafka

For scalability and durability.

crux-kafka-embedded

Useful for experimentation and testing.

Local KV and Standalone mode storage

crux-lmdb

Better read performance for intensive querying.

crux-rocksdb

Better write performance for heavy ingestion.

crux-rocksdb is a good default choice.

HTTP

Server

Client

Additional

crux-rdf

Import RDF data and run a subset of SPARQL queries.

Uberjar

crux-uberjar

One dependency to rule them all.

Crux Labs

Have a look at the more experimental modules being developed over at Crux Labs, such as crux-dataflow and the crux-console.

We will use Crux Labs as a place to test ideas and early module development, that may eventually be officially supported and migrated to the official Crux repository.

Using Clojure

Please note that Clojure is not required when using Crux. HTTP and Java APIs are also available.

REPL

Launch a REPL using the very latest Clojars -SNAPSHOT release:

clj -Sdeps '{:deps {juxt/crux-core {:mvn/version "RELEASE"}}}'

Start a standalone in-memory (i.e. not persisted anywhere) node:

(require '[crux.api :as crux])
(import '[crux.api ICruxAPI])

(def my-node
  (crux/start-node
    ;; see 'configuration' section of docs for LMDB/RocksDB storage options
    {:crux.node/topology ['crux.standalone/topology]}))

put a document:

(def my-document
  {:crux.db/id :some/fancy-id
   :arbitrary-key ["an untyped value" 123]
   :nested-map {"and values" :can-be-arbitrarily-nested}})

(crux/submit-tx my-node [[:crux.tx/put my-document]])

Take an immutable snapshot of the database:

(def my-db (crux/db my-node))

Retrieve the current version of the document:

(crux/entity my-db :some/fancy-id)

Developing Crux

• To run a REPL that includes dependencies for all components of Crux, first build the sub-modules using lein sub install.
• Start a REPL with lein repl (with --headless if you're just going to connect to it from your editor).
• Once you've connected to the REPL, in the user namespace, run:
  • (go) to start up the dev node
  • (halt!) to stop it
  • (reset) to stop it, reload changed namespaces, and restart it
  • (reset-all) to stop it, reload all namespaces, and restart it
  • if you're using Emacs/CIDER, cider-ns-refresh will do all this for you - C-c M-n M-r, , s x in Spacemacs
  • Conjure users can use ConjureRefresh, see the docs for bindings
  • see Integrant REPL for more details.
• You should now have a running Crux node under (:node user/system) - you can verify this by calling (crux/status (:node system)).
• Most of the time, you shouldn't need to bounce the REPL, but:
  • if you add a module, or change any of the dependencies of any of the modules, that'll require another lein sub install and a REPL bounce
  • if you change any of the Java classes, that'll require a lein sub javac and a REPL bounce
  • otherwise, (user/reset) (or just (reset) if you're already in the user ns) should be sufficient.
• You can run module tests from the root of the git repo without a lein sub install, because of the lein checkouts - all of the tests are in scope here, so things like lein test :only crux.tx-test should also work.
• Please don't put any more side-effecting top-level code in dev namespaces - you'll break this reload ability and make me sad.

Testing

The recommended way of running the full test suite is lein build.

Copyright & License

The MIT License (MIT)

Copyright © 2018-2019 JUXT LTD.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Dependencies

A list of compiled dependencies and corresponding licenses is available here.