https://hugegraph.apache.org/versions/1.0/docs/quickstart/computing/Recent content in Graph computing on Apache HugeGraphHugoen-USMon, 01 Jan 0001 00:00:00 +0000https://hugegraph.apache.org/versions/1.0/docs/quickstart/computing/hugegraph-computer/Mon, 01 Jan 0001 00:00:00 +0000https://hugegraph.apache.org/versions/1.0/docs/quickstart/computing/hugegraph-computer/<h2 id="1-hugegraph-computer-overview">1 HugeGraph-Computer Overview </h2> <p>The <a href="https://github.com/apache/incubator-hugegraph-computer"><code>HugeGraph-Computer</code></a> is a distributed graph processing system for HugeGraph (OLAP). It is an implementation of <a href="https://kowshik.github.io/JPregel/pregel_paper.pdf">Pregel</a>. It runs on Kubernetes framework.</p> <h3 id="features">Features </h3> <ul> <li>Support distributed MPP graph computing, and integrates with HugeGraph as graph input/output storage.</li> <li>Based on BSP(Bulk Synchronous Parallel) model, an algorithm performs computing through multiple parallel iterations, every iteration is a superstep.</li> <li>Auto memory management. The framework will never be OOM(Out of Memory) since it will split some data to disk if it doesn&rsquo;t have enough memory to hold all the data.</li> <li>The part of edges or the messages of super node can be in memory, so you will never lose it.</li> <li>You can load the data from HDFS or HugeGraph, or any other system.</li> <li>You can output the results to HDFS or HugeGraph, or any other system.</li> <li>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.</li> </ul> <h2 id="2-get-started">2 Get Started </h2> <h3 id="21-run-pagerank-algorithm-locally">2.1 Run PageRank algorithm locally </h3> <blockquote> <p>To run algorithm with HugeGraph-Computer, you need to install 64-bit JRE/JDK 11 or later versions.</p>