Skip to content
The self-tuning database for ACID transactions and ad hoc analytics across time.
Java PHP Ruby Python Thrift Shell Other
Latest commit 9aa4ace Jun 16, 2016 @jtnelson jtnelson committed on GitHub Merge pull request #178 from cinchapi/feature/CON-471
CON-471: Removed Engine#browse() method
Failed to load latest commit information.
concourse-bash-tests CON-172: add bats unit test to verify that relative paths can be impo… Oct 23, 2015
concourse-cli make cli creation more intuitive with respect to args Mar 2, 2016
concourse-driver-java Add a Reflection util to get Methods with unboxed params Jun 16, 2016
concourse-driver-php CON-470: Update param order in driver docs as per GH-115 Jun 16, 2016
concourse-driver-python CON-470: Update param order in driver docs as per GH-115 Jun 16, 2016
concourse-driver-ruby CON-470: Update param order in driver docs as per GH-115 Jun 16, 2016
concourse-ete-test-core Fix bug in call functionality of ReflectiveClient Jun 16, 2016
concourse-import CON-465: Trim keys and values when importing Apr 30, 2016
concourse-integration-tests Merge pull request #174 from cinchapi/feature/CON-469 Jun 15, 2016
concourse-plugin-core add additional constructors and refactor logic Mar 29, 2016
concourse-server CON-471: Removed Engine#browse() method Jun 16, 2016
concourse-shell IntelliJ support: launch config Jun 9, 2016
concourse-unit-test-core update copyright year to 2016 Jan 2, 2016
concourse-upgrade-tests update copyright year to 2016 Jan 2, 2016
documentation/markdown CON-158: add cashdocs for reconcile Jun 5, 2016
examples update copyright year to 2016 Jan 2, 2016
git-hooks add script to sync versioned git hooks Oct 23, 2015
gradle/wrapper upgrade to gradle 2.8 Dec 8, 2015
impl-notes Update Nov 23, 2015
interface Fix typo Jun 2, 2016
licenses Add dep on jansi-1.11 and accent4j-1.0.0 Dec 10, 2015
makeself Generate install and upgrade scripts during build Dec 18, 2013
mockcourse Implement reconcileKeyRecordValues in Mockcourse.groovy Jun 2, 2016
utils add tools to generate rest api docs for concourse server Mar 26, 2016
vagrant source the right file for centos vagrant box Oct 26, 2015
.gitignore add support for intellij Jun 9, 2016
.version bumping version to 0.5.0 Jun 17, 2014 update changelog Jan 8, 2016 Update Oct 21, 2015
LICENSE update license note about 3rd party deps Nov 12, 2015
NOTICE update copyright year to 2016 Jan 2, 2016 Update Jun 13, 2016
build.gradle add support for intellij Jun 9, 2016
gradlew allow init to happen from any directory Oct 24, 2015
gradlew.bat Added gradle wrapper to repo Oct 6, 2013 Update Oct 31, 2015
settings.gradle delete concourse-java-core project Oct 30, 2015 add license header to update-copyright Oct 28, 2015 doc update May 25, 2016 Remove filtering of error meessages cause thats not the real issue Jan 7, 2016


Join the chat at

Concourse is a self-tuning database that enables live analytics for large streams of operational data at scale. Developers use Concourse to quickly build software that requires both ACID transactions and the ability to get data insights on demand. With Concourse, end-to-end data management requires no extra infrastructure, no prior configuration and no additional coding–all of which greatly reduce costs and allow developers to focus on core business problems.

This is version 0.5.0 of Concourse.


Let's assume we have the an array of JSON objects that describe NBA players.

from concourse import *

data = [
        "name": "Lebron James",
        "age": 30,
        "team": "Cleveland Cavaliers"
        "name": "Kevin Durant",
        "age": 26,
        "team": "OKC Thunder"
        "name": "Kobe Bryant",
        "age": 36,
        "team": "LA Lakers"
You can use Concourse to quickly insert the data and do some quick analysis. Notice that we don't have to declare a schema, create any structure or configure any indexes.
concourse = Concourse.connect()
records = concourse.insert(data) # each object is added to a distinct record
lebron = records[0]
durant = records[1]
kobe = records[2]
You can read and modify individual attributes without loading the entire record.
concourse.get(key="age", record=kobe)
concourse.add(key="name", value="KD", record=durant)
concourse.remove(key="jersey_number", value=23, record=lebron)
You can easily find records that match a criteria and select the desired since everything is automatically indexed."team = Chicago Bulls")["name", "team"], criteria="age bw 22 29")
You can even query data from the past without doing any extra work.
concourse.get(key="age", record=durant, time="04/2009")
concourse.find("team = Chicago Bulls at 2011")"age > 25 and team != Chicago Bulls", time="two years ago")
It is very easy to analyze how data has changed over time and revert to previous states.
# Analyze how data has changed over time and revert to previous states
concourse.audit(key="team", record=lebron)
concourse.revert(key="jersey_number", record=kobe, time="two years ago")
And ACID transactions are available for important, cross record changes.
    concourse.set(key="current_team", value="OKC Thunder", record=lebron)
    concourse.set(key="current_team", value="Cleveland Cavs", record=durant)
except TransactionException:

You can find more examples in the examples directory. More information is also available in the Concourse Guide and API documentation.


Whether you use SQL or NoSQL, it's hard to get real-time insight into your mission critical data because most systems are only optimized for either transactions or analytics, not both. As a result, end-to-end data management requires complex data pipelining, which slows down development, complicates infrastructure and increases costs.

The Solution

Concourse is an integrated and self-managing transactional database that enables real time ad-hoc analytics without any configuration.

Automatic Indexing

You no longer need to plan queries in advance because Concourse automatically indexes all of your data while guaranteeing constant time writes that are super fast. Concourse fully supports ad-hoc range and predicate queries and automatically caches frequently requested data for optimal performance.

Version Control

Concourse automatically tracks changes to your data, just like Git does for source code. Of course this means that you can easily audit changes and revert to previous states without downtime; but it also means that you have the power to query data from the past. Version control in Concourse makes it possible to build applications that know what was known when and can analyze real-time changes over time.

ACID Transactions

Concourse supports truly distributed ACID transactions without restriction. And we use dynamic resource management with just-in-time locking to ensure that they deliver both the highest performance and strongest consistency. So no need to guess when your data will eventually become consistent. When distributed Concourse responds to a query, you can trust the results immediately.

Simple Data Model

Concourse's document-graph structure is lightweight and flexible–it supports any kind of data and very large scales. Data about each person, place or thing is stored in a record, which is simply a collection of key/value pairs. And you can create links among records to easily model all the relationships within your data.


Since Concourse makes very few assumptions about data, it integrates with your application seamlessly and never needs a translator (goodbye object-relational impedance mismatch)! You never have to declare structure up front–no schema, tables, or indexes–or specify value types because Concourse is smart enough to figure it out. Concourse dynamically adapts to your application so that you can focus on building value without having to drag the database along.


Concourse supports rich full text search right out the box, so you don't need to deploy an external search server. Data is automatically indexed and searchable in real-time without ever diminishing write performance. In Concourse, you can always perform as-you-type searches that match full or partial terms.


System Requirements
  • At least 256 MB of available memory
  • Linux or OS X
  • Java 1.7+

Concourse will be maintained under the Semantic Versioning guidelines such that release versions will be formatted as <major>.<minor>.<patch> where

  • breaking backward compatibility bumps the major,
  • new additions while maintaining backward compatibility bumps the minor, and
  • bug fixes or miscellaneous changes bumps the patch.
Additional Resources


Read the contributing guidelines to learn how to get involved in the community. We value and welcome constructive contributions from anyone, regardless of skill level :)

Mailing Lists


  • Jeff Nelson

Copyright © 2013-2016 Cinchapi Inc.

Concourse is released under the Apache License, Version 2.0. For more information see LICENSE, which is included with this package.

Something went wrong with that request. Please try again.