Migrate project to computer directory

.github/workflows/ci.yml

@@ -6,6 +6,9 @@ on:
       - master
       - /^release-.*$/
   pull_request:
+defaults:
+  run:
+    working-directory: computer
 
 jobs:
   computer-ci:
@@ -32,6 +35,7 @@ jobs:
 
       - name: License check(RAT)
         run: |
+          pwd
           mvn apache-rat:check -ntp
           find ./ -name rat.txt -print0 | xargs -0 -I file cat file > merged-rat.txt
           grep "Binaries" merged-rat.txt -C 3 && cat merged-rat.txt
@@ -87,7 +91,7 @@ jobs:
         if: ${{ env.USE_STAGE == 'true' }}
         run: |
           cp $HOME/.m2/settings.xml /tmp/settings.xml
-          mv -vf .github/configs/settings.xml $HOME/.m2/settings.xml
+          mv -vf ../.github/configs/settings.xml $HOME/.m2/settings.xml
 
       - name: Compile
         run: mvn clean compile -e -Dmaven.javadoc.skip=true -ntp

.github/workflows/codeql-analysis.yml

@@ -10,6 +10,10 @@ on:
 #  schedule:
 #    - cron: '45 7 * * 1'
 
+defaults:
+  run:
+    working-directory: computer
+
 jobs:
   analyze:
     env:
@@ -40,7 +44,7 @@ jobs:
         if: ${{ env.USE_STAGE == 'true' }}
         run: |
           cp $HOME/.m2/settings.xml /tmp/settings.xml
-          mv -vf .github/configs/settings.xml $HOME/.m2/settings.xml
+          mv -vf ../.github/configs/settings.xml $HOME/.m2/settings.xml
 
       # Initializes the CodeQL tools for scanning.
       - name: Initialize CodeQL

.github/workflows/license-checker.yml

@@ -7,6 +7,10 @@ on:
       - /^release-.*$/
   pull_request:
 
+defaults:
+  run:
+    working-directory: computer
+
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
   cancel-in-progress: true
@@ -53,7 +57,7 @@ jobs:
         if: ${{ env.USE_STAGE == 'true' }}
         run: |
           cp $HOME/.m2/settings.xml /tmp/settings.xml
-          mv -vf .github/configs/settings.xml $HOME/.m2/settings.xml
+          mv -vf ../.github/configs/settings.xml $HOME/.m2/settings.xml
       - name: Compile install
         run: mvn install -DskipTests=true -ntp
       - name: Generate & check current 3rd-party dependencies

.gitignore

@@ -32,7 +32,6 @@ dist.sh
 /nbbuild/
 /nbdist/
 /.nb-gradle/
-dist/
 build/
 
 ### VS Code ###
@@ -58,7 +57,7 @@ build/
 *.pyc
 
 # maven ignore
-output/
+
 apache-hugegraph-*-incubating-*/
 *.war
 *.zip

README.md

@@ -5,16 +5,7 @@
 [![codecov](https://codecov.io/gh/apache/incubator-hugegraph-computer/branch/master/graph/badge.svg)](https://codecov.io/gh/apache/incubator-hugegraph-computer)
 [![Docker Pulls](https://img.shields.io/docker/pulls/hugegraph/hugegraph-computer)](https://hub.docker.com/repository/docker/hugegraph/hugegraph-computer)
 
-The hugegraph-computer is a distributed graph processing system for hugegraph. It is an implementation of [Pregel](https://kowshik.github.io/JPregel/pregel_paper.pdf). It runs on Kubernetes or YARN framework.
-
-## Features
-
-- Support distributed MPP graph computing, and integrates with HugeGraph as graph input/output storage.
-- Based on BSP(Bulk Synchronous Parallel) model, an algorithm performs computing through multiple parallel iterations, every iteration is a superstep.
-- Auto memory management. The framework will never be OOM(Out of Memory) since it will split some data to disk if it doesn't have enough memory to hold all the data.
-- The part of edges or the messages of super node can be in memory, so you will never lose it.
-- You can load the data from HDFS or HugeGraph, output the results to HDFS or HugeGraph, or adapt any other systems manually as needed.
-- Easy to develop a new algorithm. You just need to focus on a vertex only processing just like as in a single server, without worrying about message transfer and memory/storage management.
+The [hugegraph-computer](./computer/README.md) is a distributed graph processing system for hugegraph.
 
 ## Learn More
 

computer/README.md

@@ -0,0 +1,12 @@
+# Apache HugeGraph-Computer
+
+The hugegraph-computer is a distributed graph processing system for hugegraph. It is an implementation of [Pregel](https://kowshik.github.io/JPregel/pregel_paper.pdf). It runs on Kubernetes or YARN framework.
+
+## Features
+
+- Support distributed MPP graph computing, and integrates with HugeGraph as graph input/output storage.
+- Based on BSP(Bulk Synchronous Parallel) model, an algorithm performs computing through multiple parallel iterations, every iteration is a superstep.
+- Auto memory management. The framework will never be OOM(Out of Memory) since it will split some data to disk if it doesn't have enough memory to hold all the data.
+- The part of edges or the messages of super node can be in memory, so you will never lose it.
+- You can load the data from HDFS or HugeGraph, output the results to HDFS or HugeGraph, or adapt any other systems manually as needed.
+- Easy to develop a new algorithm. You just need to focus on a vertex only processing just like as in a single server, without worrying about message transfer and memory/storage management.

computer-core/src/main/java/org/apache/hugegraph/computer/core/bsp/Bsp4Master.java → computer/computer-core/src/main/java/org/apache/hugegraph/computer/core/bsp/Bsp4Master.java

@@ -54,7 +54,7 @@ public List<ContainerInfo> waitWorkersInitDone() {
         LOG.info("Master is waiting for workers init-done");
         String path = this.constructPath(BspEvent.BSP_WORKER_INIT_DONE);
         List<byte[]> serializedContainers = this.waitOnWorkersEvent(
-                                            path, this.registerTimeout());
+                path, this.registerTimeout());
         List<ContainerInfo> containers = new ArrayList<>(this.workerCount());
         for (byte[] serializedContainer : serializedContainers) {
             ContainerInfo container = new ContainerInfo();
@@ -110,7 +110,7 @@ public List<WorkerStat> waitWorkersStepDone(int superstep) {
         String path = this.constructPath(BspEvent.BSP_WORKER_STEP_DONE,
                                          superstep);
         List<byte[]> list = this.waitOnWorkersEvent(path,
-                            this.barrierOnWorkersTimeout());
+                                                    this.barrierOnWorkersTimeout());
         List<WorkerStat> result = new ArrayList<>(this.workerCount());
         for (byte[] bytes : list) {
             WorkerStat workerStat = new WorkerStat();

computer-core/src/main/java/org/apache/hugegraph/computer/core/receiver/MessageRecvManager.java → computer/computer-core/src/main/java/org/apache/hugegraph/computer/core/receiver/MessageRecvManager.java

@@ -18,9 +18,9 @@
 package org.apache.hugegraph.computer.core.receiver;
 
 import java.util.Map;
+import java.util.concurrent.CompletableFuture;
 import java.util.concurrent.ExecutionException;
 import java.util.concurrent.TimeUnit;
-import java.util.concurrent.CompletableFuture;
 import java.util.concurrent.TimeoutException;
 import java.util.concurrent.atomic.AtomicInteger;
 

computer-core/src/main/java/org/apache/hugegraph/computer/core/receiver/MessageRecvPartitions.java → computer/computer-core/src/main/java/org/apache/hugegraph/computer/core/receiver/MessageRecvPartitions.java

@@ -17,6 +17,11 @@
 
 package org.apache.hugegraph.computer.core.receiver;
 
+import java.io.File;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+
 import org.apache.commons.io.FileUtils;
 import org.apache.hugegraph.computer.core.common.ComputerContext;
 import org.apache.hugegraph.computer.core.config.Config;
@@ -27,11 +32,6 @@
 import org.apache.hugegraph.computer.core.store.SuperstepFileGenerator;
 import org.apache.hugegraph.computer.core.store.entry.KvEntry;
 
-import java.io.File;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-
 public abstract class MessageRecvPartitions<P extends MessageRecvPartition> {
 
     protected final ComputerContext context;