This is an automated email from the ASF dual-hosted git repository.
lzljs3620320 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/paimon-website.git
The following commit(s) were added to refs/heads/master by this push:
new fa2127798c feat: add release-1.1
fa2127798c is described below
commit fa2127798ccea0703c061034230649d5d2dcecc8
Author: JingsongLi <[email protected]>
AuthorDate: Fri May 16 13:19:41 2025 +0800
feat: add release-1.1
---
angular.json | 3 +-
community/docs/releases/release-1.0.md | 1 +
community/docs/releases/release-1.1.md | 80 +++++++++++++++++++++++++++++++++
public/img/1.1-postpone.png | Bin 0 -> 223276 bytes
public/img/1.1-rest-api.png | Bin 0 -> 631983 bytes
5 files changed, 83 insertions(+), 1 deletion(-)
diff --git a/angular.json b/angular.json
index 72e2770614..311f331272 100644
--- a/angular.json
+++ b/angular.json
@@ -135,6 +135,7 @@
"cli": {
"schematicCollections": [
"@angular-eslint/schematics"
- ]
+ ],
+ "analytics": "27d076df-e79a-494a-a4b5-82bc0a9e9a5c"
}
}
diff --git a/community/docs/releases/release-1.0.md
b/community/docs/releases/release-1.0.md
index 5aaceb27cd..77d9a0f536 100644
--- a/community/docs/releases/release-1.0.md
+++ b/community/docs/releases/release-1.0.md
@@ -2,6 +2,7 @@
title: "Release 1.0"
type: release
version: 1.0.1
+weight: 92
---
# Apache Paimon 1.0 Available
diff --git a/community/docs/releases/release-1.1.md
b/community/docs/releases/release-1.1.md
new file mode 100644
index 0000000000..9074b55598
--- /dev/null
+++ b/community/docs/releases/release-1.1.md
@@ -0,0 +1,80 @@
+---
+title: "Release 1.1"
+type: release
+version: 1.1.1
+weight: 93
+---
+
+# Apache Paimon 1.1 Available
+
+May 16, 2025 - Jingsong Lee ([email protected])
+
+Apache Paimon PMC Officially Releases Milestone 1.1 Stable Version (Version
1.1.1).
+
+This core version has been carefully polished over nearly four months,
aggregating the wisdom of over 60 developers from the
+global open-source community. In total, it has accumulated over 400 code
commits, fully demonstrating the power of community-driven
+technological evolution. We sincerely express our gratitude to all
contributors!
+
+## Highlights
+
+1. REST Catalog: Provides a lightweight implementation to access Catalog
services, completely decoupling dependency on Hive Metastore. Designed
specifically for Paimon's unique features.
+2. External File Mode: Enabled via data-file.external-paths, allowing files to
exist outside table directories. Unlocks support for super-large tables and
partitioned files migration to cold storage.
+3. Adaptive Bucketing: Configuring bucket = -2 activates Postpone Bucket Mode,
designed to address challenges in bucket definition. Empowers backend services
to perform intelligent compaction and rescaling.
+4. File System Ecosystem: Adds support for Tencent Cloud Object Storage,
Microsoft Azure Storage, and Huawei OBS file systems.
+5. Compatibility Notes: When file-index.bitmap.columns is enabled, Bitmap
Index V2 is activated by default. Older readers cannot recognize this format;
users must upgrade to new readers if using new writers.
+
+## REST Catalog
+
+REST Catalog is ready in this release, benefits of REST Catalog:
+1. Decoupled Architecture
+ - Interacts with Catalog Server via well-defined REST APIs.
+ - Enables independent evolution and extension of Catalog Server and REST
clients.
+2. Language Agnosticism
+ - Developers can implement Catalog Server in any programming language as
long as REST APIs are followed.
+ - Leverages existing team expertise and tech stacks.
+3. Vendor Lock-in Avoidance
+ - Designed to work with any Catalog Server.
+ - Reduces migration costs for REST servers, as clients only need to comply
with REST protocols.
+
+<img src="./img/1.1-rest-api.png" alt="rest" />
+
+REST API definition
(https://paimon.apache.org/docs/master/concepts/rest/rest-api/):
+
+## External Files
+
+Apache Paimon's external files were previously stored entirely within the
table folder in earlier versions. This allowed the Manifest to record only
relative paths for files,
+offering the following advantages:
+
+1. Smaller Manifest size
+2. Easier migration (simply copy files directly)
+
+However, some requirements necessitate storing files externally, such as
managing cold/hot data across multiple file systems. This release introduces
the following parameters to address these needs:
+
+1. data-file.external-paths: Comma-separated external paths where table data
will be written.
+2. data-file.external-paths.strategy: Strategy for selecting external paths
when writing data:
+ - none mode: No external storage is selected; data remains in the default
warehouse path.
+ - specific-fs: Choose a specific file system as the external path (e.g.,
S3, OSS).
+ - round-robin: Sequentially select paths from data-file.external-paths when
writing new files.
+
+If certain data can be cooled, you can use these configurations to copy
specific partitions to object storage to reduce costs.
+
+## Adaptive Bucket (Postpone Bucket)
+
+Bucket configuration has long been a key challenge in Paimon usage, and
Compaction management is also complex. The Adaptive Bucket (Postpone Bucket)
mode was designed to simplify these complexities.
+
+<img src="./img/1.1-postpone.png" alt="rest" />
+
+There is a temporary area for storing fresh data, providing backend services
with flexible control space. This allows users to avoid worrying about Bucket
configuration and Compaction stability.
+
+## Other Features
+
+1. This version supports in-place upgrade from Iceberg tables to Paimon
tables, which is suitable for HDFS-like file systems.
+2. The 'precommit-compact' parameter is now supported for non-keyed tables,
significantly reducing small files during streaming writes.
+3. A new parameter dynamic-bucket.max-buckets has been introduced to prevent
excessive Bucket generation in keyed tables when using bucket = -1 mode.
+4. The default value of sync-all-properties for HiveCatalog tables has been
set to True, ensuring all Paimon table options are synchronized to Hive
Metastore (HMS) by default.
+
+## Future Roadmap
+
+1. Develop a Light REST SDK with dependency-free design to avoid introducing
Hadoop-related dependencies and reduce usage costs.
+2. Expand Iceberg Snapshots compatibility to support real-time data
integration with Iceberg’s Deletion Vectors, currently focusing on top-level
files in keyed tables.
+3. Enable file cloning from Hive, Hudi, and Iceberg tables to Paimon tables.
diff --git a/public/img/1.1-postpone.png b/public/img/1.1-postpone.png
new file mode 100644
index 0000000000..374f18d321
Binary files /dev/null and b/public/img/1.1-postpone.png differ
diff --git a/public/img/1.1-rest-api.png b/public/img/1.1-rest-api.png
new file mode 100644
index 0000000000..cb911ae91a
Binary files /dev/null and b/public/img/1.1-rest-api.png differ
