This Confluence has been LDAP enabled, if you are an ASF Committer, please use your LDAP Credentials to login. Any problems file an INFRA jira ticket please.

Page tree
Skip to end of metadata
Go to start of metadata

This document is a community maintained roadmap of the upcoming releases and features (at a high level).
The document contains only estimations, no guaranteed feature / release dates. 

See also the time-based release model.


Last update: May 2016

Next scheduled major release: 1.4.0
Feature freeze (branch forking):  4. September 2017
Code freeze (first voting RC):  18 September 2017
Release date29 September 2017


ReleaseDateTime span (days)



26 Aug 2014 
0.704 Nov 201470
0.821 Jan 201578
0.924 Jun 2015154
0.1016 Nov 2015145
1.008 Mar 2016113
1.108 Aug 2016153


2 Feb 2017178
1.326 May 2017119
1.429 September 2017 
1.526 January 2018 


Past releases

Flink 1.1 (August 2016)

  • Session Windows and Allowed lateness
  • Table API and SQL on Calcite
  • Metrics system
  • Scala API for CEP
  • Graph generators and new Gelly library algorithms
  • Continuous File Sources
  • Kinesis Connector
  • Cassandra Connector

Flink 1.2

Flink 1.3

  • (Large) State Handling / Recovery:

  • Streaming APIs:

  • Deployment & Tooling:

    • Flink History Server (FLINK-1579)

    • Flink automatically manages TM network buffer (FLINK-4545)

    • Add Datadog HTTP metrics reporter (FLINK-6013)

    • Add watermark monitoring to JobManager web frontend (FLINK-3427)

  • Table API / SQL:

    • Support for Retraction in Table API / SQL (FLINK-6047)

    • Support for non-windowed aggregations in streaming Table API / SQL (FLINK-6216)
    • Support for append, upsert and retraction StreamTableSinks (FLINK-6093)
    • Integrate time indicators for Table API & SQL (FLINK-5884)

    • Support for group windows in (batch + streaming) SQL (TUMBLE, HOP, SESSION) (FLINK-6261,FLINK-6011)
    • Support for OVER windows in streaming Table API and SQL (FLINK-4557)
    • Support for QueryConfig for streaming queries to specify state retention interval (FLINK-6491)
    • Support for user-defined aggregation functions (FLINK-5564)
    • Support for standard deviation and variance aggregation functions (FLINK-4604)
    • Support for TableSources with filter push down (FLINK-3849,

    • Support for external catalogs (FLINK-5570)

    • Add KafkaAvroTableSource (FLINK-3871)
    • Add HBaseTableSource (FLINK-2168)

    • Add CUBE/ROLLUP/GROUPING SETS operator in batch SQL (FLINK-5303)

    • Support for DISTINCT GROUP BY aggregates for batch SQL (FLINK-3849)
    • Add flink-table.jar to ./opt folder (FLINK-6247)
    • Support for more scalar functions (SIN, COS, TAN, DEGREES, PI etc.) (FLINK-6112)
  • Connectors:
    • Elasticsearch 5.x support (FLINK-4988)

    • Implement rescalable non-partitioned state for Kinesis Connector (FLINK-4821)

    • Transparent shard discovery for Kinesis Consumer (FLINK-4577)
    • Allow Kinesis Consumer to start from specific timestamp / Date (FLINK-4523)

    • Allow setting custom start-offsets for the Kafka consumer (FLINK-3123)

    • Allow out-opt from offset committing for the Kafka consumer (FLINK-3398)

  • CEP library improvements:

  • Gelly Library:
    • Unified driver for running Gelly examples (FLINK-4949)
    • PageRank algorithm for directed graphs (FLINK-4896)

    • Add Circulant and Echo graph generators (FLINK-6393)

Upcoming releases

Flink 1.4

Flink 1.5




  • No labels