From 152be37c1669ef4b98e63e42b9653e23499ce633 Mon Sep 17 00:00:00 2001
From: Zhen Wang <643348094@qq.com>
Date: Mon, 9 Dec 2024 14:08:34 +0800
Subject: [PATCH] [CORE][UNIFFLE] Bump uniffle 0.9.1 (#8166)

* Bump uniffle 0.9.1

* address comment

* address comment
---
 .github/workflows/velox_backend.yml | 23 ++++++++++-------------
 docs/get-started/Velox.md           |  2 +-
 pom.xml                             |  2 +-
 tools/gluten-it/pom.xml             |  2 +-
 4 files changed, 13 insertions(+), 16 deletions(-)

diff --git a/.github/workflows/velox_backend.yml b/.github/workflows/velox_backend.yml
index ce195c4e1b52..5feb5fa56855 100644
--- a/.github/workflows/velox_backend.yml
+++ b/.github/workflows/velox_backend.yml
@@ -491,6 +491,8 @@ jobs:
       fail-fast: false
       matrix:
         spark: [ "spark-3.2" ]
+        uniffle: [ "0.9.1" ]
+        hadoop: [ "2.8.5" ]
     runs-on: ubuntu-20.04
     container: centos:8
     steps:
@@ -513,21 +515,16 @@ jobs:
         run: |
           yum update -y && yum install -y java-1.8.0-openjdk-devel wget git
           $SETUP install_maven
-      - name: Build for Uniffle 0.9.0
+      - name: Install Uniffle ${{ matrix.uniffle }}
         run: |
           export JAVA_HOME=/usr/lib/jvm/java-1.8.0-openjdk && \
           cd /opt && \
-          git clone -b v0.9.0 https://github.com/apache/incubator-uniffle.git && \
-          cd incubator-uniffle && \
-          $MVN_CMD clean install -Phadoop2.8,spark3 -DskipTests
-          cd /opt && \
-          ${WGET_CMD} https://archive.apache.org/dist/incubator/uniffle/0.9.0/apache-uniffle-0.9.0-incubating-bin.tar.gz && \
-          tar xzf apache-uniffle-0.9.0-incubating-bin.tar.gz -C /opt/ && mv /opt/rss-0.9.0-hadoop2.8 /opt/uniffle && \
-          ${WGET_CMD} https://archive.apache.org/dist/hadoop/common/hadoop-2.8.5/hadoop-2.8.5.tar.gz && \
-          tar xzf hadoop-2.8.5.tar.gz -C /opt/
-          rm -rf /opt/incubator-uniffle
+          ${WGET_CMD} https://archive.apache.org/dist/incubator/uniffle/${{ matrix.uniffle }}/apache-uniffle-${{ matrix.uniffle }}-incubating-bin.tar.gz && \
+          mkdir /opt/uniffle && tar xzf apache-uniffle-${{ matrix.uniffle }}-incubating-bin.tar.gz -C /opt/uniffle --strip-components=1 && \
+          ${WGET_CMD} https://archive.apache.org/dist/hadoop/common/hadoop-${{ matrix.hadoop }}/hadoop-${{ matrix.hadoop }}.tar.gz && \
+          tar xzf hadoop-${{ matrix.hadoop }}.tar.gz -C /opt/ && \
           cd /opt/uniffle && mkdir shuffle_data && \
-          bash -c "echo -e 'XMX_SIZE=16g\nHADOOP_HOME=/opt/hadoop-2.8.5' > ./bin/rss-env.sh" && \
+          bash -c "echo -e 'XMX_SIZE=16g\nHADOOP_HOME=/opt/hadoop-${{ matrix.hadoop }}' > ./bin/rss-env.sh" && \
           bash -c "echo -e 'rss.coordinator.shuffle.nodes.max 1\nrss.rpc.server.port 19999' > ./conf/coordinator.conf" && \
           bash -c "echo -e 'rss.server.app.expired.withoutHeartbeat 7200000\nrss.server.heartbeat.delay 3000\nrss.rpc.server.port 19997\nrss.rpc.server.type GRPC_NETTY\nrss.jetty.http.port 19996\nrss.server.netty.port 19995\nrss.storage.basePath /opt/uniffle/shuffle_data\nrss.storage.type MEMORY_LOCALFILE\nrss.coordinator.quorum localhost:19999\nrss.server.flush.thread.alive 10\nrss.server.single.buffer.flush.threshold 64m' > ./conf/server.conf" && \
           bash ./bin/start-coordinator.sh && bash ./bin/start-shuffle-server.sh
@@ -535,11 +532,11 @@ jobs:
         run: |
           cd $GITHUB_WORKSPACE/ && \
           $MVN_CMD clean install -P${{ matrix.spark }} -Pbackends-velox -Puniffle -DskipTests
-      - name: TPC-H SF1.0 && TPC-DS SF1.0 Parquet local spark3.2 with uniffle-0.9.0
+      - name: TPC-H SF1.0 && TPC-DS SF1.0 Parquet local spark3.2 with uniffle-${{ matrix.uniffle }}
         run: |
           export JAVA_HOME=/usr/lib/jvm/java-1.8.0-openjdk && \
           cd $GITHUB_WORKSPACE/tools/gluten-it && \
-          $MVN_CMD clean install -Pspark-3.2 -Puniffle  && \
+          $MVN_CMD clean install -P${{ matrix.spark }} -Puniffle  && \
           GLUTEN_IT_JVM_ARGS=-Xmx5G sbin/gluten-it.sh queries-compare \
             --local --preset=velox-with-uniffle --benchmark-type=h --error-on-memleak --off-heap-size=10g -s=1.0 --threads=16 --iterations=1
 
diff --git a/docs/get-started/Velox.md b/docs/get-started/Velox.md
index 3b3ab07a8dea..a0af0cc75d02 100644
--- a/docs/get-started/Velox.md
+++ b/docs/get-started/Velox.md
@@ -272,7 +272,7 @@ spark.dynamicAllocation.enabled false
 
 ## Uniffle support
 
-Uniffle with velox backend supports [Uniffle](https://github.com/apache/incubator-uniffle) as remote shuffle service. Currently, the supported Uniffle versions are `0.9.0`.
+Uniffle with velox backend supports [Uniffle](https://github.com/apache/incubator-uniffle) as remote shuffle service. Currently, the supported Uniffle versions are `0.9.1`.
 
 First refer to this URL(https://uniffle.apache.org/docs/intro) to get start with uniffle.
 
diff --git a/pom.xml b/pom.xml
index a2cccda1e45e..9b27a6839712 100644
--- a/pom.xml
+++ b/pom.xml
@@ -66,7 +66,7 @@
     <delta.version>2.4.0</delta.version>
     <delta.binary.version>24</delta.binary.version>
     <celeborn.version>0.5.2</celeborn.version>
-    <uniffle.version>0.9.0</uniffle.version>
+    <uniffle.version>0.9.1</uniffle.version>
     <arrow.version>15.0.0</arrow.version>
     <arrow-gluten.version>15.0.0-gluten</arrow-gluten.version>
     <arrow-memory.artifact>arrow-memory-unsafe</arrow-memory.artifact>
diff --git a/tools/gluten-it/pom.xml b/tools/gluten-it/pom.xml
index 1973a7d99890..eb2c3d87e761 100644
--- a/tools/gluten-it/pom.xml
+++ b/tools/gluten-it/pom.xml
@@ -22,7 +22,7 @@
     <scala.binary.version>2.12</scala.binary.version>
     <spark.major.version>3</spark.major.version>
     <celeborn.version>0.3.2-incubating</celeborn.version>
-    <uniffle.version>0.9.0</uniffle.version>
+    <uniffle.version>0.9.1</uniffle.version>
     <gluten.version>1.3.0-SNAPSHOT</gluten.version>
     <guava.version>32.0.1-jre</guava.version>
     <tpch.version>1.1</tpch.version>