You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@flink.apache.org by ma...@apache.org on 2022/04/19 08:13:07 UTC

[flink-web] branch asf-site updated (16607c8e4 -> 6e8fb6ece)

This is an automated email from the ASF dual-hosted git repository.

martijnvisser pushed a change to branch asf-site
in repository https://gitbox.apache.org/repos/asf/flink-web.git


    from 16607c8e4 [FLINK-25897] Update gradle quickstart to gradle 7.3.3
     new 9d03051f9 [FLINK-26446] Update Feature Radar in Apache Flink Roadmap for the 1.15 release
     new 6e8fb6ece Rebuild website

The 2 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails.  The revisions
listed as "add" were already present in the repository and have only
been added to this reference.


Summary of changes:
 content/img/flink_feature_radar_3.svg    |   4 +
 content/img/flink_feature_radar_zh_3.svg |   4 +
 content/q/gradle-quickstart.sh           |  71 ++++----
 content/roadmap.html                     | 173 +++++--------------
 content/zh/roadmap.html                  | 286 +++++++++++--------------------
 img/flink_feature_radar_3.svg            |   4 +
 img/flink_feature_radar_zh_3.svg         |   4 +
 roadmap.md                               | 136 ++++-----------
 roadmap.zh.md                            | 237 ++++++++++---------------
 9 files changed, 324 insertions(+), 595 deletions(-)
 create mode 100644 content/img/flink_feature_radar_3.svg
 create mode 100644 content/img/flink_feature_radar_zh_3.svg
 create mode 100644 img/flink_feature_radar_3.svg
 create mode 100644 img/flink_feature_radar_zh_3.svg


[flink-web] 02/02: Rebuild website

Posted by ma...@apache.org.
This is an automated email from the ASF dual-hosted git repository.

martijnvisser pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/flink-web.git

commit 6e8fb6ece4ca9cfe5dc0adab93352d1654ed195c
Author: MartijnVisser <ma...@apache.org>
AuthorDate: Tue Apr 19 10:11:53 2022 +0200

    Rebuild website
---
 content/img/flink_feature_radar_3.svg    |   4 +
 content/img/flink_feature_radar_zh_3.svg |   4 +
 content/q/gradle-quickstart.sh           |  71 ++++----
 content/roadmap.html                     | 173 +++++--------------
 content/zh/roadmap.html                  | 286 +++++++++++--------------------
 5 files changed, 186 insertions(+), 352 deletions(-)

diff --git a/content/img/flink_feature_radar_3.svg b/content/img/flink_feature_radar_3.svg
new file mode 100644
index 000000000..08e5e4ad8
--- /dev/null
+++ b/content/img/flink_feature_radar_3.svg
@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!-- Do not edit this file with editors other than diagrams.net -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" width="1461px" height="1611px" viewBox="-0.5 -0.5 1461 1611" content="&lt;mxfile host=&quot;app.diagrams.net&quot; modified=&quot;2022-04-19T07:50:28.330Z&quot; agent=&quot;5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.3 Safari/605.1.15&quot; etag=&quot;4SBsAjeUeQcoES03DVG8&quot; version=&quot;17.4.3&quot; type=&quot;device&quot;&gt;&lt;diagram id [...]
\ No newline at end of file
diff --git a/content/img/flink_feature_radar_zh_3.svg b/content/img/flink_feature_radar_zh_3.svg
new file mode 100644
index 000000000..08e5e4ad8
--- /dev/null
+++ b/content/img/flink_feature_radar_zh_3.svg
@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!-- Do not edit this file with editors other than diagrams.net -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" width="1461px" height="1611px" viewBox="-0.5 -0.5 1461 1611" content="&lt;mxfile host=&quot;app.diagrams.net&quot; modified=&quot;2022-04-19T07:50:28.330Z&quot; agent=&quot;5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.3 Safari/605.1.15&quot; etag=&quot;4SBsAjeUeQcoES03DVG8&quot; version=&quot;17.4.3&quot; type=&quot;device&quot;&gt;&lt;diagram id [...]
\ No newline at end of file
diff --git a/content/q/gradle-quickstart.sh b/content/q/gradle-quickstart.sh
index b977c69e2..bb17a40ab 100755
--- a/content/q/gradle-quickstart.sh
+++ b/content/q/gradle-quickstart.sh
@@ -42,7 +42,10 @@ defaultProjectName="quickstart"
 defaultOrganization="org.myorg.quickstart"
 defaultVersion="0.1-SNAPSHOT"
 defaultFlinkVersion="${1:-1.14.4}"
-defaultScalaBinaryVersion="${2:-2.11}"
+# flink-docs-master/docs/dev/datastream/project-configuration/#gradle
+# passes the scala version prefixed with a _, e.g.: _2.12
+scalaBinaryVersionFromCmdArg="${2/_/}"
+defaultScalaBinaryVersion="${scalaBinaryVersionFromCmdArg:-2.12}"
 
 echo "This script creates a Flink project using Java and Gradle."
 
@@ -100,53 +103,49 @@ rootProject.name = '${projectName}'
 EOF
 
 cat > build.gradle <<EOF
-buildscript {
-    repositories {
-        jcenter() // this applies only to the Gradle 'Shadow' plugin
-    }
-    dependencies {
-        classpath 'com.github.jengelman.gradle.plugins:shadow:2.0.4'
-    }
-}
-
 plugins {
     id 'java'
     id 'application'
     // shadow plugin to produce fat JARs
-    id 'com.github.johnrengelman.shadow' version '2.0.4'
+    id 'com.github.johnrengelman.shadow' version '7.1.2'
 }
 
-
-// artifact properties
-group = '${organization}'
-version = '${version}'
-mainClassName = '${organization}.StreamingJob'
-description = """Flink Quickstart Job"""
-
 ext {
     javaVersion = '1.8'
     flinkVersion = '${flinkVersion}'
     scalaBinaryVersion = '${scalaBinaryVersion}'
-    slf4jVersion = '1.7.7'
-    log4jVersion = '1.2.17'
+    slf4jVersion = '1.7.32'
+    log4jVersion = '2.17.1'
+    flinkVersionNew = flinkVersion.toString().replace("-SNAPSHOT", "") >= "1.15"
 }
 
+// artifact properties
+group = '${organization}'
+version = '${version}'
+if (flinkVersionNew) {
+    mainClassName = '${organization}.DataStreamJob'
+} else {
+    mainClassName = '${organization}.StreamingJob'
+}
+description = """Flink Quickstart Job"""
 
 sourceCompatibility = javaVersion
 targetCompatibility = javaVersion
 tasks.withType(JavaCompile) {
-	options.encoding = 'UTF-8'
+    options.encoding = 'UTF-8'
 }
 
-applicationDefaultJvmArgs = ["-Dlog4j.configuration=log4j.properties"]
-
-task wrapper(type: Wrapper) {
-    gradleVersion = '3.1'
-}
+applicationDefaultJvmArgs = ["-Dlog4j.configurationFile=log4j2.properties"]
 
 // declare where to find the dependencies of your project
 repositories {
-    mavenCentral()$( if [[ "${flinkVersion}" == *-SNAPSHOT ]] ; then echo -e "\n    maven { url \"https://repository.apache.org/content/repositories/snapshots/\" }" ; else echo ""; fi )
+    mavenCentral()
+    maven {
+        url "https://repository.apache.org/content/repositories/snapshots"
+        mavenContent {
+            snapshotsOnly()
+        }
+    }
 }
 
 // NOTE: We cannot use "compileOnly" or "shadow" configurations since then we could not run code
@@ -160,7 +159,7 @@ configurations {
     flinkShadowJar.exclude group: 'org.apache.flink', module: 'force-shading'
     flinkShadowJar.exclude group: 'com.google.code.findbugs', module: 'jsr305'
     flinkShadowJar.exclude group: 'org.slf4j'
-    flinkShadowJar.exclude group: 'log4j'
+    flinkShadowJar.exclude group: 'org.apache.logging.log4j'
 }
 
 // declare the dependencies for your production and test code
@@ -169,17 +168,23 @@ dependencies {
     // Compile-time dependencies that should NOT be part of the
     // shadow jar and are provided in the lib folder of Flink
     // --------------------------------------------------------------
-    compile "org.apache.flink:flink-java:\${flinkVersion}"
-    compile "org.apache.flink:flink-streaming-java_\${scalaBinaryVersion}:\${flinkVersion}"
+    if (flinkVersionNew) {
+        implementation "org.apache.flink:flink-streaming-java:\${flinkVersion}"
+        implementation "org.apache.flink:flink-clients:\${flinkVersion}"
+    } else {
+        implementation "org.apache.flink:flink-streaming-java_\${scalaBinaryVersion}:\${flinkVersion}"
+        implementation "org.apache.flink:flink-clients_\${scalaBinaryVersion}:\${flinkVersion}"
+    }
 
     // --------------------------------------------------------------
     // Dependencies that should be part of the shadow jar, e.g.
     // connectors. These must be in the flinkShadowJar configuration!
     // --------------------------------------------------------------
-    //flinkShadowJar "org.apache.flink:flink-connector-kafka-0.11_\${scalaBinaryVersion}:\${flinkVersion}"
+    //flinkShadowJar "org.apache.flink:flink-connector-kafka:\${flinkVersion}"
 
-    compile "log4j:log4j:\${log4jVersion}"
-    compile "org.slf4j:slf4j-log4j12:\${slf4jVersion}"
+    runtimeOnly "org.apache.logging.log4j:log4j-slf4j-impl:\${log4jVersion}"
+    runtimeOnly "org.apache.logging.log4j:log4j-api:\${log4jVersion}"
+    runtimeOnly "org.apache.logging.log4j:log4j-core:\${log4jVersion}"
 
     // Add test dependencies here.
     // testCompile "junit:junit:4.12"
diff --git a/content/roadmap.html b/content/roadmap.html
index 1b7e7786c..7dd8314b8 100644
--- a/content/roadmap.html
+++ b/content/roadmap.html
@@ -258,18 +258,15 @@ under the License.
   <li><a href="#unified-analytics-where-batch-and-streaming-come-together-sql-and-beyond" id="markdown-toc-unified-analytics-where-batch-and-streaming-come-together-sql-and-beyond">Unified Analytics: Where Batch and Streaming come Together; SQL and beyond.</a>    <ul>
       <li><a href="#a-unified-sql-platform" id="markdown-toc-a-unified-sql-platform">A unified SQL Platform</a></li>
       <li><a href="#deep-batch--streaming-unification-for-the-datastream-api" id="markdown-toc-deep-batch--streaming-unification-for-the-datastream-api">Deep Batch / Streaming Unification for the DataStream API</a></li>
-      <li><a href="#subsuming-dataset-with-datastream-and-table-api" id="markdown-toc-subsuming-dataset-with-datastream-and-table-api">Subsuming DataSet with DataStream and Table API</a></li>
     </ul>
   </li>
   <li><a href="#applications-vs-clusters-flink-as-a-library" id="markdown-toc-applications-vs-clusters-flink-as-a-library">Applications vs. Clusters; “Flink as a Library”</a></li>
   <li><a href="#performance" id="markdown-toc-performance">Performance</a>    <ul>
       <li><a href="#faster-checkpoints-and-recovery" id="markdown-toc-faster-checkpoints-and-recovery">Faster Checkpoints and Recovery</a></li>
-      <li><a href="#large-scale-batch-applications" id="markdown-toc-large-scale-batch-applications">Large Scale Batch Applications</a></li>
     </ul>
   </li>
-  <li><a href="#python-apis" id="markdown-toc-python-apis">Python APIs</a></li>
+  <li><a href="#apache-flink-as-part-of-an-ever-evolving-data-ecosystem" id="markdown-toc-apache-flink-as-part-of-an-ever-evolving-data-ecosystem">Apache Flink as part of an ever evolving data ecosystem</a></li>
   <li><a href="#documentation" id="markdown-toc-documentation">Documentation</a></li>
-  <li><a href="#miscellaneous-operational-tools" id="markdown-toc-miscellaneous-operational-tools">Miscellaneous Operational Tools</a></li>
   <li><a href="#stateful-functions" id="markdown-toc-stateful-functions">Stateful Functions</a></li>
 </ul>
 
@@ -298,7 +295,7 @@ are approaching end-of-life. For questions, please contact the developer mailing
 <a href="&#109;&#097;&#105;&#108;&#116;&#111;:&#100;&#101;&#118;&#064;&#102;&#108;&#105;&#110;&#107;&#046;&#097;&#112;&#097;&#099;&#104;&#101;&#046;&#111;&#114;&#103;">&#100;&#101;&#118;&#064;&#102;&#108;&#105;&#110;&#107;&#046;&#097;&#112;&#097;&#099;&#104;&#101;&#046;&#111;&#114;&#103;</a></p>
 
 <div class="row front-graphic">
-  <img src="/img/flink_feature_radar_2.svg" width="700px" />
+  <img src="/img/flink_feature_radar_3.svg" width="700px" />
 </div>
 
 <h2 id="feature-stages">Feature Stages</h2>
@@ -335,39 +332,34 @@ platform, and is continuing to do so.</p>
 and as continuous queries. Flink already contains an efficient unified query engine, and a wide set of
 integrations. With user feedback, those are continuously improved.</p>
 
-<p><strong>More Connector and Change Data Capture Support</strong></p>
+<p><strong>Going Beyond a SQL Stream/Batch Processing Engine</strong></p>
 
 <ul>
-  <li>Change-Data-Capture: Capturing a stream of data changes, directly from databases, by attaching to the
-transaction log. The community is adding more CDC intrgrations.
-    <ul>
-      <li>External CDC connectors: <a href="https://flink-packages.org/packages/cdc-connectors">https://flink-packages.org/packages/cdc-connectors</a></li>
-      <li>Background: <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=147427289">FLIP-105</a>
-(CDC support for SQL) and <a href="https://debezium.io/">Debezium</a>.</li>
-    </ul>
-  </li>
-  <li>Data Lake Connectors: Unified streaming &amp; batch is a powerful value proposition for Data Lakes: supporting
-same APIs, semantics, and engine for streaming real-time processing and batch processing of historic data.
-The community is adding deeper integrations with various Data Lake systems:
-    <ul>
-      <li><a href="https://iceberg.apache.org/">Apache Iceberg</a>: <a href="https://iceberg.apache.org/flink/">https://iceberg.apache.org/flink/</a></li>
-      <li><a href="https://hudi.apache.org/">Apache Hudi</a>: <a href="https://hudi.apache.org/blog/apache-hudi-meets-apache-flink/">https://hudi.apache.org/blog/apache-hudi-meets-apache-flink/</a></li>
-      <li><a href="https://pinot.apache.org/">Apache Pinot</a>: <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=177045634">FLIP-166</a></li>
-    </ul>
-  </li>
+  <li>To extend the capability of a pure stream processor and make Flink ready for future use cases, 
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-188%3A+Introduce+Built-in+Dynamic+Table+Storage">FLIP-188</a>
+has been announced adding built in dynamic table storage.</li>
+  <li>The experience of updating Flink SQL based jobs has been rather cumbersome as it could have
+lead to new job graphs making restoring from savepoints/checkpoints impossible. 
+<a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=191336489&amp;src=contextnavpagetreemode">FLIP-190</a>
+that already has been shipped as MVP is targeting this.</li>
 </ul>
 
 <p><strong>Platform Infrastructure</strong></p>
 
 <ul>
-  <li>To simplify the building of production SQL platforms with Flink, we are improving the SQL client and are
-working on SQL gateway components that interface between client and cluster: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-163%3A+SQL+Client+Improvements">FLIP-163</a></li>
+  <li>After <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-163%3A+SQL+Client+Improvements">FLIP-163</a> 
+the community is working again on a set of SQL Client usability improvements
+(<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-189%3A+SQL+Client+Usability+Improvements">FLIP-189</a>)
+which is aiming at improving the user experience, when using the SQL client.</li>
 </ul>
 
 <p><strong>Support for Common Languages, Formats, Catalogs</strong></p>
 
 <ul>
-  <li>Hive Query Compatibility: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-152%3A+Hive+Query+Syntax+Compatibility">FLIP-152</a></li>
+  <li>With <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-216%3A++Introduce+pluggable+dialect+and++decouple+Hive+connector">FLIP-216</a>
+there’s now the initiative to introduce pluggable dialects on the example of the Hive connector.
+Including so many dependencies to make dialects work has lead to an overhead for contributors 
+and users.</li>
 </ul>
 
 <p>Flink has a broad SQL coverage for batch (full TPC-DS support) and a state-of-the-art set of supported
@@ -408,59 +400,6 @@ API in the process. See <a href="https://cwiki.apache.org/confluence/display/FLI
   </li>
 </ul>
 
-<p><strong>DataStream Batch Execution</strong></p>
-
-<ul>
-  <li>
-    <p>Flink is adding a <em>batch execution mode</em> for bounded DataStream programs. This gives users faster and simpler
-execution and recovery of their bounded streaming applications; users do not need to worry about watermarks and
-state sizes in this execution mode: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-140%3A+Introduce+batch-style+execution+for+bounded+keyed+streams">FLIP-140</a></p>
-
-    <p>The core batch execution mode is implemented with <a href="https://flink.apache.org/news/2020/12/10/release-1.12.0.html#batch-execution-mode-in-the-datastream-api">great results</a>;
-there are ongoing improvements around aspects like broadcast state and processing-time-timers.
-This mode requires the new unified sources and sinks that are mentioned above, so it is limited
-to the connectors that have been ported to those new APIs.</p>
-  </li>
-</ul>
-
-<p><strong>Mixing bounded/unbounded streams, and batch/streaming execution</strong></p>
-
-<ul>
-  <li>
-    <p>Support checkpointing when some tasks finished &amp; Bounded stream programs shut down with a final
-checkpoint: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-147%3A+Support+Checkpoints+After+Tasks+Finished">FLIP-147</a></p>
-  </li>
-  <li>
-    <p>There are initial discussions and designs about jobs with mixed batch/streaming execution, so stay tuned for more
-news in that area.</p>
-  </li>
-</ul>
-
-<h2 id="subsuming-dataset-with-datastream-and-table-api">Subsuming DataSet with DataStream and Table API</h2>
-
-<p>We want to eventually drop the legacy Batch-only DataSet API, have batch-and stream processing unified
-throughout the entire system.</p>
-
-<p>Overall Discussion: <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=158866741">FLIP-131</a></p>
-
-<p>The <em>DataStream API</em> supports batch-execution to efficiently execute streaming programs on historic data
-(see above). Takes over that set of use cases.</p>
-
-<p>The <em>Table API</em> should become the default API for batch-only applications.</p>
-
-<ul>
-  <li>Add more operations to Table API, so support common data manipulation tasks more
-   easily: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-155%3A+Introduce+a+few+convenient+operations+in+Table+API">FLIP-155</a></li>
-  <li>Make Source and Sink definitions easier in the Table API.</li>
-</ul>
-
-<p>Improve the <em>interplay between the Table API and the DataStream API</em> to allow switching from Table API to
-DataStream API when more control over the data types and operations is necessary.</p>
-
-<ul>
-  <li>Interoperability between DataStream and Table APIs: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-136%3A++Improve+interoperability+between+DataStream+and+Table+API">FLIP-136</a></li>
-</ul>
-
 <hr />
 
 <h1 id="applications-vs-clusters-flink-as-a-library">Applications vs. Clusters; “Flink as a Library”</h1>
@@ -471,19 +410,9 @@ job as a self contained application.</p>
 
 <p>For example as a simple Kubernetes deployment; deployed and scaled like a regular application without extra workflows.</p>
 
-<p>Deploy Flink jobs as self-contained Applications works for all deployment targets since Flink 1.11.0
-(<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-85+Flink+Application+Mode">FLIP-85</a>).</p>
-
 <ul>
-  <li>
-    <p>Reactive Scaling lets Flink applications change their parallelism in response to growing and shrinking
-worker pools, and makes Flink compatibel with standard auto-scalers:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-159%3A+Reactive+Mode">FLIP-159</a></p>
-  </li>
-  <li>
-    <p>Kubernetes-based HA-services let Flink applications run on Kubernetes without requiring a ZooKeeper dependency:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-144%3A+Native+Kubernetes+HA+for+Flink">FLIP-144</a></p>
-  </li>
+  <li>There is currently a Kubernetes Operator being developed by the community. See
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-212%3A+Introduce+Flink+Kubernetes+Operator">FLIP-212</a>.</li>
 </ul>
 
 <hr />
@@ -499,45 +428,31 @@ Checkpoints and recovery are stable and have been a reliable workhorse for years
 trying to make it faster, more predictable, and to remove some confusions and inflexibility in some areas.</p>
 
 <ul>
-  <li>Unaligned Checkpoints, to make checkpoints progress faster when applications cause backpressure:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-76%3A+Unaligned+Checkpoints">FLIP-76</a>, available
-since Flink 1.12.2.</li>
-  <li>Log-based Checkpoints, for very frequent incremental checkpointing:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-158%3A+Generalized+incremental+checkpoints">FLIP-158</a></li>
-</ul>
-
-<h2 id="large-scale-batch-applications">Large Scale Batch Applications</h2>
-
-<p>The community is working on making large scale batch execution (parallelism in the order of 10,000s)
-simpler (less configuration tuning required) and more performant.</p>
-
-<ul>
-  <li>
-    <p>Introduce a more scalable batch shuffle. First parts of this have been merged, and ongoing efforts are
-to make the memory footprint (JVM direct memory) more predictable, see
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-148%3A+Introduce+Sort-Merge+Based+Blocking+Shuffle+to+Flink">FLIP-148</a></p>
-
-    <ul>
-      <li><a href="https://issues.apache.org/jira/browse/FLINK-20740">FLINK-20740</a></li>
-      <li><a href="https://issues.apache.org/jira/browse/FLINK-19938">FLINK-19938</a></li>
-    </ul>
-  </li>
-  <li>
-    <p>Make scheduler faster for higher parallelism: <a href="https://issues.apache.org/jira/browse/FLINK-21110">FLINK-21110</a></p>
-  </li>
+  <li><a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-183%3A+Dynamic+buffer+size+adjustment">FLIP-183</a>
+is targeting size of checkpoints by debloating the buffers. A first beta is already available.</li>
+  <li>With <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-151%3A+Incremental+snapshots+for+heap-based+state+backend">FLIP-151</a> 
+there is an ongoing effort to implement a heap based state backend.</li>
 </ul>
 
 <hr />
 
-<h1 id="python-apis">Python APIs</h1>
+<h1 id="apache-flink-as-part-of-an-ever-evolving-data-ecosystem">Apache Flink as part of an ever evolving data ecosystem</h1>
 
-<p>Most functionalities in the Java Table APIs and DataStream APIs are already supported by the Python APIs. 
-The community is continuously working on improvements such as improving the checkpoint strategy for Python UDF execution
-(<a href="https://issues.apache.org/jira/browse/FLINK-18235">FLINK-18235</a>), introducing more connectors support in both the Python DataStream API 
-and Python Table API so that the Python API can be used in for production implementations.</p>
+<p>There is almost no use case in which Apache Flink is used on its own. It has established itself
+as part of many data related reference architectures. In fact you’ll find the squirrel logo covering
+several aspects. The community has added a lot of connectors and formats. With the already mentionend
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface">FLIP-27</a> and
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API">FLIP-143</a>
+a new default for connectors has been established.</p>
 
-<p>Stateful transformation functions for the Python DataStream API:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-153%3A+Support+state+access+in+Python+DataStream+API">FLIP-153</a></p>
+<ul>
+  <li>There are efforts to revise the formats API with
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-219%3A+Revised+Format+API">FLIP-219</a></li>
+  <li>There is ongoing work on new connectors 
+(e.g. <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=177045634">Pinot</a>)</li>
+  <li>Connectors will be hosted in an external repository going forward. See the 
+<a href="https://lists.apache.org/thread/8k1xonqt7hn0xldbky1cxfx3fzh6sj7h">ML thread</a></li>
+</ul>
 
 <hr />
 
@@ -554,16 +469,6 @@ of the documentation.</p>
 
 <hr />
 
-<h1 id="miscellaneous-operational-tools">Miscellaneous Operational Tools</h1>
-
-<ul>
-  <li>Allow switching state backends with savepoints: <a href="https://issues.apache.org/jira/browse/FLINK-20976">FLINK-20976</a></li>
-  <li>Support for Savepoints with more properties, like incremental savepoints, etc.:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-47%3A+Checkpoints+vs.+Savepoints">FLIP-47</a></li>
-</ul>
-
-<hr />
-
 <h1 id="stateful-functions">Stateful Functions</h1>
 
 <p>The Stateful Functions subproject has its own roadmap published under <a href="https://statefun.io/">statefun.io</a>.</p>
diff --git a/content/zh/roadmap.html b/content/zh/roadmap.html
index a54a1a0be..99e378f49 100644
--- a/content/zh/roadmap.html
+++ b/content/zh/roadmap.html
@@ -253,21 +253,18 @@ under the License.
       <li><a href="#section-1" id="markdown-toc-section-1">功能阶段</a></li>
     </ul>
   </li>
-  <li><a href="#sql" id="markdown-toc-sql">一体化分析:流批一体,SQL及其他</a>    <ul>
-      <li><a href="#sql-1" id="markdown-toc-sql-1">统一的SQL平台</a></li>
-      <li><a href="#datastream-api-" id="markdown-toc-datastream-api-">DataStream API 流批一体深度融合</a></li>
-      <li><a href="#datastream--table-api--dataset" id="markdown-toc-datastream--table-api--dataset">使用 DataStream &amp; Table API 取代 DataSet</a></li>
+  <li><a href="#unified-analytics-where-batch-and-streaming-come-together-sql-and-beyond" id="markdown-toc-unified-analytics-where-batch-and-streaming-come-together-sql-and-beyond">Unified Analytics: Where Batch and Streaming come Together; SQL and beyond.</a>    <ul>
+      <li><a href="#a-unified-sql-platform" id="markdown-toc-a-unified-sql-platform">A unified SQL Platform</a></li>
+      <li><a href="#deep-batch--streaming-unification-for-the-datastream-api" id="markdown-toc-deep-batch--streaming-unification-for-the-datastream-api">Deep Batch / Streaming Unification for the DataStream API</a></li>
     </ul>
   </li>
   <li><a href="#applications-vs-clusters-flink-as-a-library" id="markdown-toc-applications-vs-clusters-flink-as-a-library">Applications vs. Clusters; “Flink as a Library”</a></li>
-  <li><a href="#section-2" id="markdown-toc-section-2">性能</a>    <ul>
+  <li><a href="#performance" id="markdown-toc-performance">Performance</a>    <ul>
       <li><a href="#faster-checkpoints-and-recovery" id="markdown-toc-faster-checkpoints-and-recovery">Faster Checkpoints and Recovery</a></li>
-      <li><a href="#section-3" id="markdown-toc-section-3">大规模批作业</a></li>
     </ul>
   </li>
-  <li><a href="#python-apis" id="markdown-toc-python-apis">Python APIs</a></li>
-  <li><a href="#section-4" id="markdown-toc-section-4">文档</a></li>
-  <li><a href="#section-5" id="markdown-toc-section-5">操作工具</a></li>
+  <li><a href="#apache-flink-as-part-of-an-ever-evolving-data-ecosystem" id="markdown-toc-apache-flink-as-part-of-an-ever-evolving-data-ecosystem">Apache Flink as part of an ever evolving data ecosystem</a></li>
+  <li><a href="#documentation" id="markdown-toc-documentation">Documentation</a></li>
   <li><a href="#stateful-functions" id="markdown-toc-stateful-functions">Stateful Functions</a></li>
 </ul>
 
@@ -295,7 +292,7 @@ under the License.
 。</p>
 
 <div class="row front-graphic">
-  <img src="/img/flink_feature_radar_zh_2.svg" width="700px" />
+  <img src="/img/flink_feature_radar_zh_3.svg" width="700px" />
 </div>
 
 <h2 id="section-1">功能阶段</h2>
@@ -311,144 +308,92 @@ under the License.
 
 <hr />
 
-<h1 id="sql">一体化分析:流批一体,SQL及其他</h1>
-<p>Flink的内核是流数据处理系统,Flink将批处理作为流的特例,用流的方式来执行批处理。
-Flink作为一个流式引擎,不仅能够高效执行批处理,更重要的是通过高效处理有限流的方式,
-打开了无缝流批一体处理之门。</p>
+<h1 id="unified-analytics-where-batch-and-streaming-come-together-sql-and-beyond">Unified Analytics: Where Batch and Streaming come Together; SQL and beyond.</h1>
 
-<p>流批一体升级了流数据范例:它可以保证实时和离线应用语义的一致性。
-此外,有时流式处理的作业也需要离线(有限流)处理作为补充,例如,
-在出现错误或出现数据质量问题时需要重新处理数据,
-或者有的情况下启动新的作业但需要历史数据作为引导。统一的API和系统使得此类操作变得很容易。</p>
+<p>Flink is a streaming data system in its core, that executes “batch as a special case of streaming”.
+Efficient execution of batch jobs is powerful in its own right; but even more so, batch processing
+capabilities (efficient processing of bounded streams) open the way for a seamless unification of
+batch and streaming applications.</p>
 
-<h2 id="sql-1">统一的SQL平台</h2>
+<p>Unified streaming/batch up-levels the streaming data paradigm: It gives users consistent semantics across
+their real-time and lag-time applications. Furthermore, streaming applications often need to be complemented
+by batch (bounded stream) processing, for example when reprocessing data after bugs or data quality issues,
+or when bootstrapping new applications. A unified API and system make this much easier.</p>
 
-<p>Flink社区一直致力于建设基于Flink的统一的流批一体SQL分析平台,并将持续在这个方向上努力。
-SQL具有非常强的跨流批的语义,并允许用户使用相同的SQL语句对即时查询(ad-hoc query)和
-连续查询(continuous query)进行分析。Flink拥有高效的统一查询引擎,以及基于此的一系列整合统一。
-根据用户反馈,我们会持续改善这些整合统一的使用体验。</p>
+<h2 id="a-unified-sql-platform">A unified SQL Platform</h2>
 
-<p><strong>CDC &amp; Connectors</strong></p>
+<p>The community has been building Flink to a powerful basis for a unified (batch and streaming) SQL analytics
+platform, and is continuing to do so.</p>
 
-<ul>
-  <li>CDC(Change-Data-Capture): 通过直接连接到数据库的日志来捕获数据变更。Flink社区会加强和CDC的整合
-    <ul>
-      <li>CDC connectors: 
-<a href="https://flink-packages.org/packages/cdc-connectors">https://flink-packages.org/packages/cdc-connectors</a></li>
-      <li>背景: <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=147427289">FLIP-105</a>
-(CDC对SQL的支持) &amp; <a href="https://debezium.io/">Debezium</a></li>
-    </ul>
-  </li>
-  <li>数据湖Connectors: 流批一体对数据湖有很大价值,包括支持流(处理当前数据)和批(处理历史数据)的相同API,相同语义,以及相同引擎。
-目前,Flink社区正在和各种数据湖系统进行深度融合,包括:
-    <ul>
-      <li><a href="https://iceberg.apache.org/">Apache Iceberg</a>: 
-<a href="https://iceberg.apache.org/flink/">https://iceberg.apache.org/flink/</a></li>
-      <li><a href="https://hudi.apache.org/">Apache Hudi</a>: 
-<a href="https://hudi.apache.org/blog/apache-hudi-meets-apache-flink/">https://hudi.apache.org/blog/apache-hudi-meets-apache-flink/</a></li>
-      <li><a href="https://pinot.apache.org/">Apache Pinot</a>: 
-<a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=177045634">FLIP-166</a></li>
-    </ul>
-  </li>
-</ul>
-
-<p><strong>SQL 平台基建</strong></p>
+<p>SQL has very strong cross-batch-streaming semantics, allowing users to use the same queries for ad-hoc analytics
+and as continuous queries. Flink already contains an efficient unified query engine, and a wide set of
+integrations. With user feedback, those are continuously improved.</p>
 
-<ul>
-  <li>为了简化Flink SQL的生产实践,我们正在改进SQL客户端以及SQL Gateway中客户端和群集之间与交互相关的组件:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-163%3A+SQL+Client+Improvements">FLIP-163</a></li>
-</ul>
-
-<p><strong>通用语言,格式(Formats),目录(Catalogs)</strong></p>
+<p><strong>Going Beyond a SQL Stream/Batch Processing Engine</strong></p>
 
 <ul>
-  <li>Hive Query兼容性支持: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-152%3A+Hive+Query+Syntax+Compatibility">FLIP-152</a></li>
+  <li>To extend the capability of a pure stream processor and make Flink ready for future use cases, 
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-188%3A+Introduce+Built-in+Dynamic+Table+Storage">FLIP-188</a>
+has been announced adding built in dynamic table storage.</li>
+  <li>The experience of updating Flink SQL based jobs has been rather cumbersome as it could have
+lead to new job graphs making restoring from savepoints/checkpoints impossible. 
+<a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=191336489&amp;src=contextnavpagetreemode">FLIP-190</a>
+that already has been shipped as MVP is targeting this.</li>
 </ul>
 
-<p>Flink SQL具备广泛的批处理覆盖(全面的TPC-DS支持)和最先进的流处理支持。我们也一直会努力增添更多的SQL功能和算子。</p>
-
-<h2 id="datastream-api-">DataStream API 流批一体深度融合</h2>
-
-<p><em>DataStream API</em> 是 Flink 的<em>物理层</em> API, 针对需要明确地控制数据类型,数据流,状态以及时间的应用。
-DataStream API 也在不断丰富演化以支持在有限数据上的高效地批处理。
-DataStream API 使用和流式一样的 dataflow 来执行批处理,并使用一样的算子。
-这样一来,用户使用 DataStream API 表达流和批可以保持相同级别的控制。
-我们对DataStream API融合的最终目标是可以混合并自由切换流批执行,提供流批间无缝切换体验。</p>
-
-<p><strong>统一 Sources &amp; Sinks</strong></p>
+<p><strong>Platform Infrastructure</strong></p>
 
 <ul>
-  <li>
-    <p>第一代 source 的 API 和实现要么只能用于 DataStream API 里面的流处理
-(<a href="https://github.com/apache/flink/blob/master/flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/source/SourceFunction.java">SourceFunction</a>)
-;要么只能用于 DataSet API 里面的批处理 
-(<a href="https://github.com/apache/flink/blob/master/flink-core/src/main/java/org/apache/flink/api/common/io/InputFormat.java">InputFormat</a>) 。</p>
-
-    <p>因此,我们致力于构造对流和批都适用的 sources,能让用户在这两种模式下有一致的使用体验,
-并可以很容易地在流处理和批处理之间切换,执行无限流和有限流作业。
-新的 Source API 的接口已经可以使用。我们会将更多的 source connectors 迁移到这个新的模型,详见
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface">FLIP-27</a>.</p>
-  </li>
-  <li>
-    <p>和 source 类似,原先的 sink 及其 API 也是分别针对流
-(<a href="https://github.com/apache/flink/blob/master/flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/sink/SinkFunction.java">SinkFunction</a>)
-和批 (<a href="https://github.com/apache/flink/blob/master/flink-core/src/main/java/org/apache/flink/api/common/io/OutputFormat.java">OutputFormat</a>)
-设计的。</p>
-
-    <p>为此,我们引入了新的 sink API,可以流批一致的解决结果输出和提交 (<em>Transactions</em>) 的问题。
-新的 sink API 的第一版已经出炉,并在不断改进中,详见 
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API">FLIP-143</a> 。</p>
-  </li>
+  <li>After <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-163%3A+SQL+Client+Improvements">FLIP-163</a> 
+the community is working again on a set of SQL Client usability improvements
+(<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-189%3A+SQL+Client+Usability+Improvements">FLIP-189</a>)
+which is aiming at improving the user experience, when using the SQL client.</li>
 </ul>
 
-<p><strong>DataStream 的批处理执行模式</strong></p>
+<p><strong>Support for Common Languages, Formats, Catalogs</strong></p>
 
 <ul>
-  <li>
-    <p>Flink 在 DataStream 上为有限流新增加了<em>批执行模式</em>,这可以使得用户更简单快速的执行和恢复有限流作业。 
-在有限流的批执行模式下,用户无需担心 watermarks 和状态大小的问题:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-140%3A+Introduce+batch-style+execution+for+bounded+keyed+streams">FLIP-140</a> 。</p>
-
-    <p>批执行模式的核心实现已有 <a href="https://flink.apache.org/news/2020/12/10/release-1.12.0.html#batch-execution-mode-in-the-datastream-api">很好的结果</a>;
-其他部分也在持续改进中,包括 broadcast state 和 processing-time-timers。
-值得注意的是,此模式的实现基于上面提到的新的的 source 和 sink,因此它只能支持已经使用新 API 的 connectors。</p>
-  </li>
+  <li>With <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-216%3A++Introduce+pluggable+dialect+and++decouple+Hive+connector">FLIP-216</a>
+there’s now the initiative to introduce pluggable dialects on the example of the Hive connector.
+Including so many dependencies to make dialects work has lead to an overhead for contributors 
+and users.</li>
 </ul>
 
-<p><strong>混合 有限流/无限流 &amp; 批执行/流执行</strong></p>
-
-<ul>
-  <li>
-    <p>支持在部分 task 结束后还可以做 checkpoint &amp; 支持有限流作业在结束的时候做最后一次 checkpoint :
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-147%3A+Support+Checkpoints+After+Tasks+Finished">FLIP-147</a></p>
-  </li>
-  <li>
-    <p>对于混合/切换流和批的执行,我们有一些初步的设计和讨论,敬请关注。</p>
-  </li>
-</ul>
+<p>Flink has a broad SQL coverage for batch (full TPC-DS support) and a state-of-the-art set of supported
+operations in streaming. There is continuous effort to add more functions and cover more SQL operations.</p>
 
-<h2 id="datastream--table-api--dataset">使用 DataStream &amp; Table API 取代 DataSet</h2>
+<h2 id="deep-batch--streaming-unification-for-the-datastream-api">Deep Batch / Streaming Unification for the DataStream API</h2>
 
-<p>我们希望最终能弃用只支持批式处理的 DataSet API,从而使用统一的流批处理贯穿整个系统。
-整体的讨论在这里: <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=158866741">FLIP-131</a> 。</p>
+<p>The <em>DataStream API</em> is Flink’s <em>physical</em> API, for use cases where users need very explicit control over data
+types, streams, state, and time. This API is evolving to support efficient batch execution on bounded data.</p>
 
-<p><em>DataStream API</em> 可以高效的用批的方式来执行需要处理历史数据的流作业(如上所述)。</p>
+<p>DataStream API executes the same dataflow shape in batch as in streaming, keeping the same operators.
+That way users keep the same level of control over the dataflow, and our goal is to mix and switch between
+batch/streaming execution in the future to make it a seamless experience.</p>
 
-<p><em>Table API</em> 应该是所有单批作业所使用的默认的 API 。</p>
+<p><strong>Unified Sources and Sinks</strong></p>
 
 <ul>
   <li>
-    <p>Table API 增加更多操作,以方便支持常见的数据操作任务
-  <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-155%3A+Introduce+a+few+convenient+operations+in+Table+API">FLIP-155</a></p>
-  </li>
-  <li>
-    <p>在 Table API 中,使 Source 和 Sink 更容易定义使用</p>
+    <p>The first APIs and implementations of sources were specific to either streaming programs in the DataStream API
+(<a href="https://github.com/apache/flink/blob/master/flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/source/SourceFunction.java">SourceFunction</a>),
+or to batch programs in the DataSet API (<a href="https://github.com/apache/flink/blob/master/flink-core/src/main/java/org/apache/flink/api/common/io/InputFormat.java">InputFormat</a>).</p>
+
+    <p>In this effort, we are creating sources that work across batch and streaming execution. The aim is to give
+users a consistent experience across both modes, and to allow them to easily switch between streaming and batch
+execution for their unbounded and bounded streaming applications.
+The interface for this New Source API is done and available, and we are working on migrating more source connectors
+to this new model, see <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface">FLIP-27</a>.</p>
   </li>
   <li>
-    <p>DataStream API &amp; Table API 互通性: 
-  <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-136%3A++Improve+interoperability+between+DataStream+and+Table+API">FLIP-136</a></p>
+    <p>Similar to the sources, the original sink APIs are also specific to streaming
+(<a href="https://github.com/apache/flink/blob/master/flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/sink/SinkFunction.java">SinkFunction</a>)
+and batch (<a href="https://github.com/apache/flink/blob/master/flink-core/src/main/java/org/apache/flink/api/common/io/OutputFormat.java">OutputFormat</a>)
+APIs and execution.</p>
 
-    <p>提升 Table API 和 DataStream API 之间互通的能力。当需要更多对数据类型和操作控制的时候,允许从 Table API 切换到 DataStream API 。</p>
+    <p>We have introduced a new API for sinks that consistently handles result writing and committing (<em>Transactions</em>)
+across batch and streaming. The first iteration of the API exists, and we are porting sinks and refining the
+API in the process. See <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API">FLIP-143</a>.</p>
   </li>
 </ul>
 
@@ -456,103 +401,74 @@ DataStream API 使用和流式一样的 dataflow 来执行批处理,并使用
 
 <h1 id="applications-vs-clusters-flink-as-a-library">Applications vs. Clusters; “Flink as a Library”</h1>
 
-<p>这个部分的工作主要是为了使部署(长时间运行的流式)Flink 作业变得更为自然简单。
-我们希望部署一个流式作业就像启动一个独立的应用(Applications)一样简单:
-不需要首先启动集群(Clusters),再向该集群提交作业。</p>
+<p>The goal of these efforts is to make it feel natural to deploy (long running streaming) Flink applications.
+Instead of starting a cluster and submitting a job to that cluster, these efforts support deploying a streaming
+job as a self contained application.</p>
 
-<p>例如,我们期望 Flink 作业可以作为简单的Kubernetes部署,能像普通应用程序一样可以进行常规部署和扩展,而无需额外的工作流程。
-从 Flink 1.11.0 开始,Flink 支持将 Flink 作业部署为独立的应用程序
-(<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-85+Flink+Application+Mode">FLIP-85</a>) 。</p>
+<p>For example as a simple Kubernetes deployment; deployed and scaled like a regular application without extra workflows.</p>
 
 <ul>
-  <li>
-    <p>响应式缩放功能(Reactive Scaling)可以使 Flink 作业根据资源池的增长和收缩情况更改并行度。 
-这样可以自然地使 Flink 与标准自动缩放(atuo Scaler)兼容。
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-159%3A+Reactive+Mode">FLIP-159</a></p>
-  </li>
-  <li>
-    <p>基于 Kubernetes 的高可用性(HA)服务使 Flink 作业在 Kubernetes 上运行时无需依赖ZooKeeper:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-144%3A+Native+Kubernetes+HA+for+Flink">FLIP-144</a></p>
-  </li>
+  <li>There is currently a Kubernetes Operator being developed by the community. See
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-212%3A+Introduce+Flink+Kubernetes+Operator">FLIP-212</a>.</li>
 </ul>
 
 <hr />
 
-<h1 id="section-2">性能</h1>
+<h1 id="performance">Performance</h1>
 
-<p>我们会持续不断的提高性能和容错恢复速度。</p>
+<p>Continuous work to keep improving performance and recovery speed.</p>
 
 <h2 id="faster-checkpoints-and-recovery">Faster Checkpoints and Recovery</h2>
 
-<p>Flink 社区正在致力于提升做检查点(checkpointing)和容错恢复(recovery)的速度。
-Flink的容错机制多年来运行非常稳定,但是我们还是想让整个容错过程更快并且更可预测,提升易用性。</p>
-
-<ul>
-  <li>
-    <p>Unaligned Checkpoints,解决反压情况下 Checkpoint 做不出来的问题,从 Flink 1.12.2 版本开始可用:
-  <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-76%3A+Unaligned+Checkpoints">FLIP-76</a></p>
-  </li>
-  <li>
-    <p>Log-based Checkpoints, 可以做高频增量 Checkpoints,加快 checkpoint:
-  <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-158%3A+Generalized+incremental+checkpoints">FLIP-158</a></p>
-  </li>
-</ul>
-
-<h2 id="section-3">大规模批作业</h2>
-
-<p>Flink 社区也在致力于简化大规模批作业(并行度量级在10,000左右)的部署运行,所需的配置调整更少并使之有更好的性能。</p>
+<p>The community is continuously working on improving checkpointing and recovery speed.
+Checkpoints and recovery are stable and have been a reliable workhorse for years. We are still
+trying to make it faster, more predictable, and to remove some confusions and inflexibility in some areas.</p>
 
 <ul>
-  <li>
-    <p>为批处理引入更具扩展性的 batch shuffle。Batch shuffle 的第一部分已经合并入社区代码,
-  剩下的部分可以使内存占用量(JVM直接内存)更可预测,请参阅
-  <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-148%3A+Introduce+Sort-Merge+Based+Blocking+Shuffle+to+Flink">FLIP-148</a></p>
-
-    <ul>
-      <li><a href="https://issues.apache.org/jira/browse/FLINK-20740">FLINK-20740</a></li>
-      <li><a href="https://issues.apache.org/jira/browse/FLINK-19938">FLINK-19938</a></li>
-    </ul>
-  </li>
-  <li>
-    <p>更快调度高并发作业:<a href="https://issues.apache.org/jira/browse/FLINK-21110">FLINK-21110</a></p>
-  </li>
+  <li><a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-183%3A+Dynamic+buffer+size+adjustment">FLIP-183</a>
+is targeting size of checkpoints by debloating the buffers. A first beta is already available.</li>
+  <li>With <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-151%3A+Incremental+snapshots+for+heap-based+state+backend">FLIP-151</a> 
+there is an ongoing effort to implement a heap based state backend.</li>
 </ul>
 
 <hr />
 
-<h1 id="python-apis">Python APIs</h1>
+<h1 id="apache-flink-as-part-of-an-ever-evolving-data-ecosystem">Apache Flink as part of an ever evolving data ecosystem</h1>
 
-<p>Python DataStream API 对状态访问的支持:
-<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-153%3A+Support+state+access+in+Python+DataStream+API">FLIP-153</a></p>
-
-<hr />
-
-<h1 id="section-4">文档</h1>
-
-<p>我们也正在简化文档结构,以方便更直观的导航和阅读</p>
+<p>There is almost no use case in which Apache Flink is used on its own. It has established itself
+as part of many data related reference architectures. In fact you’ll find the squirrel logo covering
+several aspects. The community has added a lot of connectors and formats. With the already mentionend
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface">FLIP-27</a> and
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API">FLIP-143</a>
+a new default for connectors has been established.</p>
 
 <ul>
-  <li>Flink 文档迁移(Jekyll to Hugo):
-  <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-157+Migrate+Flink+Documentation+from+Jekyll+to+Hugo">FLIP-157</a></li>
-  <li>文档重构: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-42%3A+Rework+Flink+Documentation">FLIP-42</a></li>
-  <li>SQL 文档: <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=127405685">FLIP-60</a></li>
+  <li>There are efforts to revise the formats API with
+<a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-219%3A+Revised+Format+API">FLIP-219</a></li>
+  <li>There is ongoing work on new connectors 
+(e.g. <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=177045634">Pinot</a>)</li>
+  <li>Connectors will be hosted in an external repository going forward. See the 
+<a href="https://lists.apache.org/thread/8k1xonqt7hn0xldbky1cxfx3fzh6sj7h">ML thread</a></li>
 </ul>
 
 <hr />
 
-<h1 id="section-5">操作工具</h1>
+<h1 id="documentation">Documentation</h1>
+
+<p>There are various dedicated efforts to simplify the maintenance and structure (more intuitive navigation/reading)
+of the documentation.</p>
 
 <ul>
-  <li>允许使用 savepoint 来切换后端状态存储(state backends): <a href="https://issues.apache.org/jira/browse/FLINK-20976">FLINK-20976</a></li>
-  <li>支持 savepoint 的其他的一些属性,例如增量 savepoint 等:
-  <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-47%3A+Checkpoints+vs.+Savepoints">FLIP-47</a></li>
+  <li>Docs Tech Stack: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-157+Migrate+Flink+Documentation+from+Jekyll+to+Hugo">FLIP-157</a></li>
+  <li>General Docs Structure: <a href="https://cwiki.apache.org/confluence/display/FLINK/FLIP-42%3A+Rework+Flink+Documentation">FLIP-42</a></li>
+  <li>SQL Docs: <a href="https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=127405685">FLIP-60</a></li>
 </ul>
 
 <hr />
 
 <h1 id="stateful-functions">Stateful Functions</h1>
 
-<p>Stateful Functions 子项目有其单独的规划路线图,请参考 <a href="https://statefun.io/">statefun.io</a> 。</p>
+<p>The Stateful Functions subproject has its own roadmap published under <a href="https://statefun.io/">statefun.io</a>.</p>
 
 
   </div>


[flink-web] 01/02: [FLINK-26446] Update Feature Radar in Apache Flink Roadmap for the 1.15 release

Posted by ma...@apache.org.
This is an automated email from the ASF dual-hosted git repository.

martijnvisser pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/flink-web.git

commit 9d03051f94a5086d8b0d9b21d9225a3738aef1de
Author: Joe Moser <jo...@ververica.com>
AuthorDate: Tue Apr 12 17:40:16 2022 +0200

    [FLINK-26446] Update Feature Radar in Apache Flink Roadmap for the 1.15 release
---
 img/flink_feature_radar_3.svg    |   4 +
 img/flink_feature_radar_zh_3.svg |   4 +
 roadmap.md                       | 136 ++++++----------------
 roadmap.zh.md                    | 237 ++++++++++++++++-----------------------
 4 files changed, 138 insertions(+), 243 deletions(-)

diff --git a/img/flink_feature_radar_3.svg b/img/flink_feature_radar_3.svg
new file mode 100644
index 000000000..08e5e4ad8
--- /dev/null
+++ b/img/flink_feature_radar_3.svg
@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!-- Do not edit this file with editors other than diagrams.net -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" width="1461px" height="1611px" viewBox="-0.5 -0.5 1461 1611" content="&lt;mxfile host=&quot;app.diagrams.net&quot; modified=&quot;2022-04-19T07:50:28.330Z&quot; agent=&quot;5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.3 Safari/605.1.15&quot; etag=&quot;4SBsAjeUeQcoES03DVG8&quot; version=&quot;17.4.3&quot; type=&quot;device&quot;&gt;&lt;diagram id [...]
\ No newline at end of file
diff --git a/img/flink_feature_radar_zh_3.svg b/img/flink_feature_radar_zh_3.svg
new file mode 100644
index 000000000..08e5e4ad8
--- /dev/null
+++ b/img/flink_feature_radar_zh_3.svg
@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!-- Do not edit this file with editors other than diagrams.net -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" width="1461px" height="1611px" viewBox="-0.5 -0.5 1461 1611" content="&lt;mxfile host=&quot;app.diagrams.net&quot; modified=&quot;2022-04-19T07:50:28.330Z&quot; agent=&quot;5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.3 Safari/605.1.15&quot; etag=&quot;4SBsAjeUeQcoES03DVG8&quot; version=&quot;17.4.3&quot; type=&quot;device&quot;&gt;&lt;diagram id [...]
\ No newline at end of file
diff --git a/roadmap.md b/roadmap.md
index bedd48fc0..edeb4ac9b 100644
--- a/roadmap.md
+++ b/roadmap.md
@@ -47,7 +47,7 @@ are approaching end-of-life. For questions, please contact the developer mailing
 [dev@flink.apache.org](mailto:dev@flink.apache.org)
 
 <div class="row front-graphic">
-  <img src="{{ site.baseurl }}/img/flink_feature_radar_2.svg" width="700px" />
+  <img src="{{ site.baseurl }}/img/flink_feature_radar_3.svg" width="700px" />
 </div>
 
 ## Feature Stages
@@ -82,29 +82,30 @@ SQL has very strong cross-batch-streaming semantics, allowing users to use the s
 and as continuous queries. Flink already contains an efficient unified query engine, and a wide set of
 integrations. With user feedback, those are continuously improved.
 
-**More Connector and Change Data Capture Support**
+**Going Beyond a SQL Stream/Batch Processing Engine**
 
-  - Change-Data-Capture: Capturing a stream of data changes, directly from databases, by attaching to the
-    transaction log. The community is adding more CDC intrgrations.
-      - External CDC connectors: [https://flink-packages.org/packages/cdc-connectors](https://flink-packages.org/packages/cdc-connectors)
-      - Background: [FLIP-105](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=147427289)
-        (CDC support for SQL) and [Debezium](https://debezium.io/).
+  - To extend the capability of a pure stream processor and make Flink ready for future use cases, 
+    [FLIP-188](https://cwiki.apache.org/confluence/display/FLINK/FLIP-188%3A+Introduce+Built-in+Dynamic+Table+Storage)
+    has been announced adding built in dynamic table storage.
+  - The experience of updating Flink SQL based jobs has been rather cumbersome as it could have
+    lead to new job graphs making restoring from savepoints/checkpoints impossible. 
+    [FLIP-190](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=191336489&src=contextnavpagetreemode)
+    that already has been shipped as MVP is targeting this.
 
-  - Data Lake Connectors: Unified streaming & batch is a powerful value proposition for Data Lakes: supporting
-    same APIs, semantics, and engine for streaming real-time processing and batch processing of historic data.
-    The community is adding deeper integrations with various Data Lake systems:
-    - [Apache Iceberg](https://iceberg.apache.org/): [https://iceberg.apache.org/flink/](https://iceberg.apache.org/flink/)
-    - [Apache Hudi](https://hudi.apache.org/): [https://hudi.apache.org/blog/apache-hudi-meets-apache-flink/](https://hudi.apache.org/blog/apache-hudi-meets-apache-flink/)
-    - [Apache Pinot](https://pinot.apache.org/): [FLIP-166](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=177045634)
 
 **Platform Infrastructure**
 
-  - To simplify the building of production SQL platforms with Flink, we are improving the SQL client and are
-    working on SQL gateway components that interface between client and cluster: [FLIP-163](https://cwiki.apache.org/confluence/display/FLINK/FLIP-163%3A+SQL+Client+Improvements)
+  - After [FLIP-163](https://cwiki.apache.org/confluence/display/FLINK/FLIP-163%3A+SQL+Client+Improvements) 
+    the community is working again on a set of SQL Client usability improvements
+    ([FLIP-189](https://cwiki.apache.org/confluence/display/FLINK/FLIP-189%3A+SQL+Client+Usability+Improvements))
+    which is aiming at improving the user experience, when using the SQL client.
 
 **Support for Common Languages, Formats, Catalogs**
 
-  - Hive Query Compatibility: [FLIP-152](https://cwiki.apache.org/confluence/display/FLINK/FLIP-152%3A+Hive+Query+Syntax+Compatibility)
+  - With [FLIP-216](https://cwiki.apache.org/confluence/display/FLINK/FLIP-216%3A++Introduce+pluggable+dialect+and++decouple+Hive+connector)
+    there's now the initiative to introduce pluggable dialects on the example of the Hive connector.
+    Including so many dependencies to make dialects work has lead to an overhead for contributors 
+    and users.
 
 Flink has a broad SQL coverage for batch (full TPC-DS support) and a state-of-the-art set of supported
 operations in streaming. There is continuous effort to add more functions and cover more SQL operations.
@@ -139,46 +140,6 @@ batch/streaming execution in the future to make it a seamless experience.
     across batch and streaming. The first iteration of the API exists, and we are porting sinks and refining the
     API in the process. See [FLIP-143](https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API).
 
-**DataStream Batch Execution**
-
-  - Flink is adding a *batch execution mode* for bounded DataStream programs. This gives users faster and simpler
-    execution and recovery of their bounded streaming applications; users do not need to worry about watermarks and
-    state sizes in this execution mode: [FLIP-140](https://cwiki.apache.org/confluence/display/FLINK/FLIP-140%3A+Introduce+batch-style+execution+for+bounded+keyed+streams)
-
-    The core batch execution mode is implemented with [great results](https://flink.apache.org/news/2020/12/10/release-1.12.0.html#batch-execution-mode-in-the-datastream-api);
-    there are ongoing improvements around aspects like broadcast state and processing-time-timers.
-    This mode requires the new unified sources and sinks that are mentioned above, so it is limited
-    to the connectors that have been ported to those new APIs.
-
-**Mixing bounded/unbounded streams, and batch/streaming execution**
-
-  - Support checkpointing when some tasks finished & Bounded stream programs shut down with a final
-    checkpoint: [FLIP-147](https://cwiki.apache.org/confluence/display/FLINK/FLIP-147%3A+Support+Checkpoints+After+Tasks+Finished)
-
-  - There are initial discussions and designs about jobs with mixed batch/streaming execution, so stay tuned for more
-    news in that area.
-
-## Subsuming DataSet with DataStream and Table API
-
-We want to eventually drop the legacy Batch-only DataSet API, have batch-and stream processing unified
-throughout the entire system.
-
-Overall Discussion: [FLIP-131](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=158866741)
-
-The _DataStream API_ supports batch-execution to efficiently execute streaming programs on historic data
-(see above). Takes over that set of use cases.
-
-The _Table API_ should become the default API for batch-only applications.
-
-  - Add more operations to Table API, so support common data manipulation tasks more
-   easily: [FLIP-155](https://cwiki.apache.org/confluence/display/FLINK/FLIP-155%3A+Introduce+a+few+convenient+operations+in+Table+API)
-  - Make Source and Sink definitions easier in the Table API.
-
-Improve the _interplay between the Table API and the DataStream API_ to allow switching from Table API to
-DataStream API when more control over the data types and operations is necessary.
-
-  - Interoperability between DataStream and Table APIs: [FLIP-136](https://cwiki.apache.org/confluence/display/FLINK/FLIP-136%3A++Improve+interoperability+between+DataStream+and+Table+API)
-
 <hr />
 
 # Applications vs. Clusters; "Flink as a Library"
@@ -189,15 +150,8 @@ job as a self contained application.
 
 For example as a simple Kubernetes deployment; deployed and scaled like a regular application without extra workflows.
 
-Deploy Flink jobs as self-contained Applications works for all deployment targets since Flink 1.11.0
-([FLIP-85](https://cwiki.apache.org/confluence/display/FLINK/FLIP-85+Flink+Application+Mode)).
-
-  - Reactive Scaling lets Flink applications change their parallelism in response to growing and shrinking
-    worker pools, and makes Flink compatibel with standard auto-scalers:
-    [FLIP-159](https://cwiki.apache.org/confluence/display/FLINK/FLIP-159%3A+Reactive+Mode)
-
-  - Kubernetes-based HA-services let Flink applications run on Kubernetes without requiring a ZooKeeper dependency:
-    [FLIP-144](https://cwiki.apache.org/confluence/display/FLINK/FLIP-144%3A+Native+Kubernetes+HA+for+Flink)
+  - There is currently a Kubernetes Operator being developed by the community. See
+    [FLIP-212](https://cwiki.apache.org/confluence/display/FLINK/FLIP-212%3A+Introduce+Flink+Kubernetes+Operator).
 
 <hr />
 
@@ -211,38 +165,28 @@ The community is continuously working on improving checkpointing and recovery sp
 Checkpoints and recovery are stable and have been a reliable workhorse for years. We are still
 trying to make it faster, more predictable, and to remove some confusions and inflexibility in some areas.
 
-  - Unaligned Checkpoints, to make checkpoints progress faster when applications cause backpressure:
-    [FLIP-76](https://cwiki.apache.org/confluence/display/FLINK/FLIP-76%3A+Unaligned+Checkpoints), available
-    since Flink 1.12.2.
-  - Log-based Checkpoints, for very frequent incremental checkpointing:
-    [FLIP-158](https://cwiki.apache.org/confluence/display/FLINK/FLIP-158%3A+Generalized+incremental+checkpoints)
-
-## Large Scale Batch Applications
-
-The community is working on making large scale batch execution (parallelism in the order of 10,000s)
-simpler (less configuration tuning required) and more performant.
-
-  - Introduce a more scalable batch shuffle. First parts of this have been merged, and ongoing efforts are
-    to make the memory footprint (JVM direct memory) more predictable, see
-    [FLIP-148](https://cwiki.apache.org/confluence/display/FLINK/FLIP-148%3A+Introduce+Sort-Merge+Based+Blocking+Shuffle+to+Flink)
-
-      - [FLINK-20740](https://issues.apache.org/jira/browse/FLINK-20740)
-      - [FLINK-19938](https://issues.apache.org/jira/browse/FLINK-19938)
-
-  - Make scheduler faster for higher parallelism: [FLINK-21110](https://issues.apache.org/jira/browse/FLINK-21110)
+  - [FLIP-183](https://cwiki.apache.org/confluence/display/FLINK/FLIP-183%3A+Dynamic+buffer+size+adjustment)
+    is targeting size of checkpoints by debloating the buffers. A first beta is already available.
+  - With [FLIP-151](https://cwiki.apache.org/confluence/display/FLINK/FLIP-151%3A+Incremental+snapshots+for+heap-based+state+backend) 
+    there is an ongoing effort to implement a heap based state backend.
 
 <hr />
 
-# Python APIs
-
-Most functionalities in the Java Table APIs and DataStream APIs are already supported by the Python APIs. 
-The community is continuously working on improvements such as improving the checkpoint strategy for Python UDF execution
-([FLINK-18235](https://issues.apache.org/jira/browse/FLINK-18235)), introducing more connectors support in both the Python DataStream API 
-and Python Table API so that the Python API can be used in for production implementations.
+# Apache Flink as part of an ever evolving data ecosystem
 
+There is almost no use case in which Apache Flink is used on its own. It has established itself
+as part of many data related reference architectures. In fact you'll find the squirrel logo covering
+several aspects. The community has added a lot of connectors and formats. With the already mentionend
+[FLIP-27](https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface) and
+[FLIP-143](https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API)
+a new default for connectors has been established.
 
-Stateful transformation functions for the Python DataStream API:
-[FLIP-153](https://cwiki.apache.org/confluence/display/FLINK/FLIP-153%3A+Support+state+access+in+Python+DataStream+API)
+  - There are efforts to revise the formats API with
+    [FLIP-219](https://cwiki.apache.org/confluence/display/FLINK/FLIP-219%3A+Revised+Format+API)
+  - There is ongoing work on new connectors 
+    (e.g. [Pinot](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=177045634))
+  - Connectors will be hosted in an external repository going forward. See the 
+    [ML thread](https://lists.apache.org/thread/8k1xonqt7hn0xldbky1cxfx3fzh6sj7h)
 
 <hr />
 
@@ -257,14 +201,6 @@ of the documentation.
 
 <hr />
 
-# Miscellaneous Operational Tools
-
-  - Allow switching state backends with savepoints: [FLINK-20976](https://issues.apache.org/jira/browse/FLINK-20976)
-  - Support for Savepoints with more properties, like incremental savepoints, etc.:
-    [FLIP-47](https://cwiki.apache.org/confluence/display/FLINK/FLIP-47%3A+Checkpoints+vs.+Savepoints)
-
-<hr />
-
 # Stateful Functions
 
 The Stateful Functions subproject has its own roadmap published under [statefun.io](https://statefun.io/).
diff --git a/roadmap.zh.md b/roadmap.zh.md
index 4b4ac1d17..9b6b833bb 100644
--- a/roadmap.zh.md
+++ b/roadmap.zh.md
@@ -46,7 +46,7 @@ under the License.
 。
 
 <div class="row front-graphic">
-  <img src="{{ site.baseurl }}/img/flink_feature_radar_zh_2.svg" width="700px" />
+  <img src="{{ site.baseurl }}/img/flink_feature_radar_zh_3.svg" width="700px" />
 </div>
 
 ## 功能阶段
@@ -61,195 +61,146 @@ under the License.
 <hr />
 
 
-# 一体化分析:流批一体,SQL及其他
-Flink的内核是流数据处理系统,Flink将批处理作为流的特例,用流的方式来执行批处理。
-Flink作为一个流式引擎,不仅能够高效执行批处理,更重要的是通过高效处理有限流的方式,
-打开了无缝流批一体处理之门。
+# Unified Analytics: Where Batch and Streaming come Together; SQL and beyond.
 
-流批一体升级了流数据范例:它可以保证实时和离线应用语义的一致性。
-此外,有时流式处理的作业也需要离线(有限流)处理作为补充,例如,
-在出现错误或出现数据质量问题时需要重新处理数据,
-或者有的情况下启动新的作业但需要历史数据作为引导。统一的API和系统使得此类操作变得很容易。
+Flink is a streaming data system in its core, that executes "batch as a special case of streaming".
+Efficient execution of batch jobs is powerful in its own right; but even more so, batch processing
+capabilities (efficient processing of bounded streams) open the way for a seamless unification of
+batch and streaming applications.
 
-## 统一的SQL平台
+Unified streaming/batch up-levels the streaming data paradigm: It gives users consistent semantics across
+their real-time and lag-time applications. Furthermore, streaming applications often need to be complemented
+by batch (bounded stream) processing, for example when reprocessing data after bugs or data quality issues,
+or when bootstrapping new applications. A unified API and system make this much easier.
 
-Flink社区一直致力于建设基于Flink的统一的流批一体SQL分析平台,并将持续在这个方向上努力。
-SQL具有非常强的跨流批的语义,并允许用户使用相同的SQL语句对即时查询(ad-hoc query)和
-连续查询(continuous query)进行分析。Flink拥有高效的统一查询引擎,以及基于此的一系列整合统一。
-根据用户反馈,我们会持续改善这些整合统一的使用体验。
+## A unified SQL Platform
 
+The community has been building Flink to a powerful basis for a unified (batch and streaming) SQL analytics
+platform, and is continuing to do so.
 
-**CDC & Connectors**
+SQL has very strong cross-batch-streaming semantics, allowing users to use the same queries for ad-hoc analytics
+and as continuous queries. Flink already contains an efficient unified query engine, and a wide set of
+integrations. With user feedback, those are continuously improved.
 
-  - CDC(Change-Data-Capture): 通过直接连接到数据库的日志来捕获数据变更。Flink社区会加强和CDC的整合
-      - CDC connectors: 
-        [https://flink-packages.org/packages/cdc-connectors](https://flink-packages.org/packages/cdc-connectors)
-      - 背景: [FLIP-105](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=147427289)
-        (CDC对SQL的支持) & [Debezium](https://debezium.io/)
+**Going Beyond a SQL Stream/Batch Processing Engine**
 
-  - 数据湖Connectors: 流批一体对数据湖有很大价值,包括支持流(处理当前数据)和批(处理历史数据)的相同API,相同语义,以及相同引擎。
-	目前,Flink社区正在和各种数据湖系统进行深度融合,包括:
-      - [Apache Iceberg](https://iceberg.apache.org/): 
-        [https://iceberg.apache.org/flink/](https://iceberg.apache.org/flink/)
-      - [Apache Hudi](https://hudi.apache.org/): 
-        [https://hudi.apache.org/blog/apache-hudi-meets-apache-flink/](https://hudi.apache.org/blog/apache-hudi-meets-apache-flink/)
-      - [Apache Pinot](https://pinot.apache.org/): 
-        [FLIP-166](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=177045634)
+  - To extend the capability of a pure stream processor and make Flink ready for future use cases, 
+    [FLIP-188](https://cwiki.apache.org/confluence/display/FLINK/FLIP-188%3A+Introduce+Built-in+Dynamic+Table+Storage)
+    has been announced adding built in dynamic table storage.
+  - The experience of updating Flink SQL based jobs has been rather cumbersome as it could have
+    lead to new job graphs making restoring from savepoints/checkpoints impossible. 
+    [FLIP-190](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=191336489&src=contextnavpagetreemode)
+    that already has been shipped as MVP is targeting this.
 
-**SQL 平台基建**
 
-  - 为了简化Flink SQL的生产实践,我们正在改进SQL客户端以及SQL Gateway中客户端和群集之间与交互相关的组件:
-    [FLIP-163](https://cwiki.apache.org/confluence/display/FLINK/FLIP-163%3A+SQL+Client+Improvements)
+**Platform Infrastructure**
 
-**通用语言,格式(Formats),目录(Catalogs)**
+  - After [FLIP-163](https://cwiki.apache.org/confluence/display/FLINK/FLIP-163%3A+SQL+Client+Improvements) 
+    the community is working again on a set of SQL Client usability improvements
+    ([FLIP-189](https://cwiki.apache.org/confluence/display/FLINK/FLIP-189%3A+SQL+Client+Usability+Improvements))
+    which is aiming at improving the user experience, when using the SQL client.
 
-  - Hive Query兼容性支持: [FLIP-152](https://cwiki.apache.org/confluence/display/FLINK/FLIP-152%3A+Hive+Query+Syntax+Compatibility)
+**Support for Common Languages, Formats, Catalogs**
 
-Flink SQL具备广泛的批处理覆盖(全面的TPC-DS支持)和最先进的流处理支持。我们也一直会努力增添更多的SQL功能和算子。
+  - With [FLIP-216](https://cwiki.apache.org/confluence/display/FLINK/FLIP-216%3A++Introduce+pluggable+dialect+and++decouple+Hive+connector)
+    there's now the initiative to introduce pluggable dialects on the example of the Hive connector.
+    Including so many dependencies to make dialects work has lead to an overhead for contributors 
+    and users.
 
+Flink has a broad SQL coverage for batch (full TPC-DS support) and a state-of-the-art set of supported
+operations in streaming. There is continuous effort to add more functions and cover more SQL operations.
 
-## DataStream API 流批一体深度融合
+## Deep Batch / Streaming Unification for the DataStream API
 
-*DataStream API* 是 Flink 的*物理层* API, 针对需要明确地控制数据类型,数据流,状态以及时间的应用。
-DataStream API 也在不断丰富演化以支持在有限数据上的高效地批处理。
-DataStream API 使用和流式一样的 dataflow 来执行批处理,并使用一样的算子。
-这样一来,用户使用 DataStream API 表达流和批可以保持相同级别的控制。
-我们对DataStream API融合的最终目标是可以混合并自由切换流批执行,提供流批间无缝切换体验。
+The *DataStream API* is Flink's *physical* API, for use cases where users need very explicit control over data
+types, streams, state, and time. This API is evolving to support efficient batch execution on bounded data.
 
-**统一 Sources & Sinks**
+DataStream API executes the same dataflow shape in batch as in streaming, keeping the same operators.
+That way users keep the same level of control over the dataflow, and our goal is to mix and switch between
+batch/streaming execution in the future to make it a seamless experience.
 
-  - 第一代 source 的 API 和实现要么只能用于 DataStream API 里面的流处理
-    ([SourceFunction](https://github.com/apache/flink/blob/master/flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/source/SourceFunction.java))
-    ;要么只能用于 DataSet API 里面的批处理 
-    ([InputFormat](https://github.com/apache/flink/blob/master/flink-core/src/main/java/org/apache/flink/api/common/io/InputFormat.java)) 。
-    
-    因此,我们致力于构造对流和批都适用的 sources,能让用户在这两种模式下有一致的使用体验,
-    并可以很容易地在流处理和批处理之间切换,执行无限流和有限流作业。
-    新的 Source API 的接口已经可以使用。我们会将更多的 source connectors 迁移到这个新的模型,详见
-	[FLIP-27](https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface).
+**Unified Sources and Sinks**
 
-  - 和 source 类似,原先的 sink 及其 API 也是分别针对流
-    ([SinkFunction](https://github.com/apache/flink/blob/master/flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/sink/SinkFunction.java))
-    和批 ([OutputFormat](https://github.com/apache/flink/blob/master/flink-core/src/main/java/org/apache/flink/api/common/io/OutputFormat.java))
-    设计的。
-
-    为此,我们引入了新的 sink API,可以流批一致的解决结果输出和提交 (*Transactions*) 的问题。
-	新的 sink API 的第一版已经出炉,并在不断改进中,详见 
-	[FLIP-143](https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API) 。
-    
-**DataStream 的批处理执行模式**
-
-  - Flink 在 DataStream 上为有限流新增加了*批执行模式*,这可以使得用户更简单快速的执行和恢复有限流作业。 
-    在有限流的批执行模式下,用户无需担心 watermarks 和状态大小的问题:
-    [FLIP-140](https://cwiki.apache.org/confluence/display/FLINK/FLIP-140%3A+Introduce+batch-style+execution+for+bounded+keyed+streams) 。
-	
-    批执行模式的核心实现已有 [很好的结果](https://flink.apache.org/news/2020/12/10/release-1.12.0.html#batch-execution-mode-in-the-datastream-api);
-    其他部分也在持续改进中,包括 broadcast state 和 processing-time-timers。
-    值得注意的是,此模式的实现基于上面提到的新的的 source 和 sink,因此它只能支持已经使用新 API 的 connectors。
-
-**混合 有限流/无限流 & 批执行/流执行**
-
-  - 支持在部分 task 结束后还可以做 checkpoint & 支持有限流作业在结束的时候做最后一次 checkpoint :
-    [FLIP-147](https://cwiki.apache.org/confluence/display/FLINK/FLIP-147%3A+Support+Checkpoints+After+Tasks+Finished)
+  - The first APIs and implementations of sources were specific to either streaming programs in the DataStream API
+    ([SourceFunction](https://github.com/apache/flink/blob/master/flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/source/SourceFunction.java)),
+    or to batch programs in the DataSet API ([InputFormat](https://github.com/apache/flink/blob/master/flink-core/src/main/java/org/apache/flink/api/common/io/InputFormat.java)).
 
-  - 对于混合/切换流和批的执行,我们有一些初步的设计和讨论,敬请关注。
+    In this effort, we are creating sources that work across batch and streaming execution. The aim is to give
+    users a consistent experience across both modes, and to allow them to easily switch between streaming and batch
+    execution for their unbounded and bounded streaming applications.
+    The interface for this New Source API is done and available, and we are working on migrating more source connectors
+    to this new model, see [FLIP-27](https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface).
 
-## 使用 DataStream & Table API 取代 DataSet
-
-我们希望最终能弃用只支持批式处理的 DataSet API,从而使用统一的流批处理贯穿整个系统。
-整体的讨论在这里: [FLIP-131](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=158866741) 。
-
-_DataStream API_ 可以高效的用批的方式来执行需要处理历史数据的流作业(如上所述)。
-
-_Table API_ 应该是所有单批作业所使用的默认的 API 。
-
-- Table API 增加更多操作,以方便支持常见的数据操作任务
-	[FLIP-155](https://cwiki.apache.org/confluence/display/FLINK/FLIP-155%3A+Introduce+a+few+convenient+operations+in+Table+API)
+  - Similar to the sources, the original sink APIs are also specific to streaming
+    ([SinkFunction](https://github.com/apache/flink/blob/master/flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/sink/SinkFunction.java))
+    and batch ([OutputFormat](https://github.com/apache/flink/blob/master/flink-core/src/main/java/org/apache/flink/api/common/io/OutputFormat.java))
+    APIs and execution.
 
-- 在 Table API 中,使 Source 和 Sink 更容易定义使用                                                  
+    We have introduced a new API for sinks that consistently handles result writing and committing (*Transactions*)
+    across batch and streaming. The first iteration of the API exists, and we are porting sinks and refining the
+    API in the process. See [FLIP-143](https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API).
 
-- DataStream API & Table API 互通性: 
-	[FLIP-136](https://cwiki.apache.org/confluence/display/FLINK/FLIP-136%3A++Improve+interoperability+between+DataStream+and+Table+API)
-	
-	提升 Table API 和 DataStream API 之间互通的能力。当需要更多对数据类型和操作控制的时候,允许从 Table API 切换到 DataStream API 。
-   
 <hr />
 
 # Applications vs. Clusters; "Flink as a Library"
 
-这个部分的工作主要是为了使部署(长时间运行的流式)Flink 作业变得更为自然简单。
-我们希望部署一个流式作业就像启动一个独立的应用(Applications)一样简单:
-不需要首先启动集群(Clusters),再向该集群提交作业。
+The goal of these efforts is to make it feel natural to deploy (long running streaming) Flink applications.
+Instead of starting a cluster and submitting a job to that cluster, these efforts support deploying a streaming
+job as a self contained application.
 
-例如,我们期望 Flink 作业可以作为简单的Kubernetes部署,能像普通应用程序一样可以进行常规部署和扩展,而无需额外的工作流程。
-从 Flink 1.11.0 开始,Flink 支持将 Flink 作业部署为独立的应用程序
-([FLIP-85](https://cwiki.apache.org/confluence/display/FLINK/FLIP-85+Flink+Application+Mode)) 。
+For example as a simple Kubernetes deployment; deployed and scaled like a regular application without extra workflows.
 
-  - 响应式缩放功能(Reactive Scaling)可以使 Flink 作业根据资源池的增长和收缩情况更改并行度。 
-    这样可以自然地使 Flink 与标准自动缩放(atuo Scaler)兼容。
-	[FLIP-159](https://cwiki.apache.org/confluence/display/FLINK/FLIP-159%3A+Reactive+Mode)
-
-  - 基于 Kubernetes 的高可用性(HA)服务使 Flink 作业在 Kubernetes 上运行时无需依赖ZooKeeper:
-    [FLIP-144](https://cwiki.apache.org/confluence/display/FLINK/FLIP-144%3A+Native+Kubernetes+HA+for+Flink)
+  - There is currently a Kubernetes Operator being developed by the community. See
+    [FLIP-212](https://cwiki.apache.org/confluence/display/FLINK/FLIP-212%3A+Introduce+Flink+Kubernetes+Operator).
 
 <hr />
 
-# 性能
+# Performance
 
-我们会持续不断的提高性能和容错恢复速度。
+Continuous work to keep improving performance and recovery speed.
 
 ## Faster Checkpoints and Recovery
 
-Flink 社区正在致力于提升做检查点(checkpointing)和容错恢复(recovery)的速度。
-Flink的容错机制多年来运行非常稳定,但是我们还是想让整个容错过程更快并且更可预测,提升易用性。
-
-- Unaligned Checkpoints,解决反压情况下 Checkpoint 做不出来的问题,从 Flink 1.12.2 版本开始可用:
-    [FLIP-76](https://cwiki.apache.org/confluence/display/FLINK/FLIP-76%3A+Unaligned+Checkpoints)
-  
-- Log-based Checkpoints, 可以做高频增量 Checkpoints,加快 checkpoint:
-	[FLIP-158](https://cwiki.apache.org/confluence/display/FLINK/FLIP-158%3A+Generalized+incremental+checkpoints)
-
-## 大规模批作业
-
-Flink 社区也在致力于简化大规模批作业(并行度量级在10,000左右)的部署运行,所需的配置调整更少并使之有更好的性能。
-
-- 为批处理引入更具扩展性的 batch shuffle。Batch shuffle 的第一部分已经合并入社区代码,
-	剩下的部分可以使内存占用量(JVM直接内存)更可预测,请参阅
-	[FLIP-148](https://cwiki.apache.org/confluence/display/FLINK/FLIP-148%3A+Introduce+Sort-Merge+Based+Blocking+Shuffle+to+Flink)
-	
-    - [FLINK-20740](https://issues.apache.org/jira/browse/FLINK-20740)
-    - [FLINK-19938](https://issues.apache.org/jira/browse/FLINK-19938)
+The community is continuously working on improving checkpointing and recovery speed.
+Checkpoints and recovery are stable and have been a reliable workhorse for years. We are still
+trying to make it faster, more predictable, and to remove some confusions and inflexibility in some areas.
 
-- 更快调度高并发作业:[FLINK-21110](https://issues.apache.org/jira/browse/FLINK-21110)
+  - [FLIP-183](https://cwiki.apache.org/confluence/display/FLINK/FLIP-183%3A+Dynamic+buffer+size+adjustment)
+    is targeting size of checkpoints by debloating the buffers. A first beta is already available.
+  - With [FLIP-151](https://cwiki.apache.org/confluence/display/FLINK/FLIP-151%3A+Incremental+snapshots+for+heap-based+state+backend) 
+    there is an ongoing effort to implement a heap based state backend.
 
 <hr />
 
-# Python APIs
+# Apache Flink as part of an ever evolving data ecosystem
 
-Python DataStream API 对状态访问的支持:
-[FLIP-153](https://cwiki.apache.org/confluence/display/FLINK/FLIP-153%3A+Support+state+access+in+Python+DataStream+API)
+There is almost no use case in which Apache Flink is used on its own. It has established itself
+as part of many data related reference architectures. In fact you'll find the squirrel logo covering
+several aspects. The community has added a lot of connectors and formats. With the already mentionend
+[FLIP-27](https://cwiki.apache.org/confluence/display/FLINK/FLIP-27%3A+Refactor+Source+Interface) and
+[FLIP-143](https://cwiki.apache.org/confluence/display/FLINK/FLIP-143%3A+Unified+Sink+API)
+a new default for connectors has been established.
 
-<hr />
-
-# 文档
-
-我们也正在简化文档结构,以方便更直观的导航和阅读
-
-- Flink 文档迁移(Jekyll to Hugo):
-	[FLIP-157](https://cwiki.apache.org/confluence/display/FLINK/FLIP-157+Migrate+Flink+Documentation+from+Jekyll+to+Hugo)
-- 文档重构: [FLIP-42](https://cwiki.apache.org/confluence/display/FLINK/FLIP-42%3A+Rework+Flink+Documentation)
-- SQL 文档: [FLIP-60](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=127405685)
+  - There are efforts to revise the formats API with
+    [FLIP-219](https://cwiki.apache.org/confluence/display/FLINK/FLIP-219%3A+Revised+Format+API)
+  - There is ongoing work on new connectors 
+    (e.g. [Pinot](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=177045634))
+  - Connectors will be hosted in an external repository going forward. See the 
+    [ML thread](https://lists.apache.org/thread/8k1xonqt7hn0xldbky1cxfx3fzh6sj7h)
 
 <hr />
 
-# 操作工具
+# Documentation
+
+There are various dedicated efforts to simplify the maintenance and structure (more intuitive navigation/reading)
+of the documentation.
 
-- 允许使用 savepoint 来切换后端状态存储(state backends): [FLINK-20976](https://issues.apache.org/jira/browse/FLINK-20976)
-- 支持 savepoint 的其他的一些属性,例如增量 savepoint 等:
-	[FLIP-47](https://cwiki.apache.org/confluence/display/FLINK/FLIP-47%3A+Checkpoints+vs.+Savepoints)
+  - Docs Tech Stack: [FLIP-157](https://cwiki.apache.org/confluence/display/FLINK/FLIP-157+Migrate+Flink+Documentation+from+Jekyll+to+Hugo)
+  - General Docs Structure: [FLIP-42](https://cwiki.apache.org/confluence/display/FLINK/FLIP-42%3A+Rework+Flink+Documentation)
+  - SQL Docs: [FLIP-60](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=127405685)
 
 <hr />
 
 # Stateful Functions
 
-Stateful Functions 子项目有其单独的规划路线图,请参考 [statefun.io](https://statefun.io/) 。
\ No newline at end of file
+The Stateful Functions subproject has its own roadmap published under [statefun.io](https://statefun.io/).
\ No newline at end of file