jerqi commented on code in PR #7333: URL: https://github.com/apache/gravitino/pull/7333#discussion_r2128102860
########## api/src/main/java/org/apache/gravitino/stats/Statistic.java: ########## @@ -0,0 +1,88 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.gravitino.stats; + +import java.util.Optional; +import org.apache.gravitino.annotation.Evolving; + +/** + * Statistic interface represents a statistic that can be associated with a metadata object. It can + * be used to store various types of statistics, for example, table statistics, partition + * statistics, fileset statistics, etc. + * + * <p>For different statistics, Gravitino will use different storage solution for them. Table + * statistics will be stored in Gravitino metadata store, usually the Gravitino's metadata is + * relational database, it's difficult to store millions rows of records. + * + * <p>But for partition statistics, there may be millions of partitions in a large table. So + * Gravitino will use a external storage solution. Gravitino will use a file to store all + * partitions. The partition statistics may be several megabytes. If a partition is stored as a + * file, it will cause millions of small files in the storage system. It will cause high pressure + * for HDFS if you choose HDFS as external storage. If the partition statistic is requested for + * multiple times, Gravitino will use cache to improve performance. + * + * <p>The file format of partition statistics is Json, Iceberg isn't design for small files. It will + * read multiple files in a single read operation. It will append many small data files if users + * update a partition every time. The relational database will record the pointer of the partition + * statistics file. Review Comment: We use Json as the solution. The following words explain why not to choose other file formats. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@gravitino.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
