{"id":1442,"date":"2020-05-28T01:59:24","date_gmt":"2020-05-27T17:59:24","guid":{"rendered":"http:\/\/39.97.231.10\/?p=1442"},"modified":"2020-05-28T01:59:24","modified_gmt":"2020-05-27T17:59:24","slug":"hudi%e9%85%8d%e7%bd%ae","status":"publish","type":"post","link":"http:\/\/www.weijing.co\/?p=1442","title":{"rendered":"Hudi\u914d\u7f6e"},"content":{"rendered":"<header><\/header>\n<section>\u672c\u6587\u4e3b\u8981\u4ecb\u7ecd\u4e86\u51e0\u79cd\u914d\u7f6e\u5199\u5165\u6216\u8bfb\u53d6Hudi\u6570\u636e\u96c6\u7684\u4f5c\u4e1a\u7684\u65b9\u6cd5\u3002 \u7b80\u800c\u8a00\u4e4b\uff0c\u60a8\u53ef\u4ee5\u5728\u51e0\u4e2a\u7ea7\u522b\u4e0a\u63a7\u5236\u884c\u4e3a\u3002<\/p>\n<ul>\n<li>Spark\u6570\u636e\u6e90\u914d\u7f6e\u00a0: \u8fd9\u4e9b\u914d\u7f6e\u63a7\u5236Hudi Spark\u6570\u636e\u6e90\uff0c\u63d0\u4f9b\u5982\u4e0b\u529f\u80fd\uff1a \u5b9a\u4e49\u952e\u548c\u5206\u533a\u3001\u9009\u62e9\u5199\u64cd\u4f5c\u3001\u6307\u5b9a\u5982\u4f55\u5408\u5e76\u8bb0\u5f55\u6216\u9009\u62e9\u8981\u8bfb\u53d6\u7684\u89c6\u56fe\u7c7b\u578b\u3002<\/li>\n<li>WriteClient \u914d\u7f6e\u00a0: \u5728\u5185\u90e8\uff0cHudi\u6570\u636e\u6e90\u4f7f\u7528\u57fa\u4e8eRDD\u7684HoodieWriteClient\u00a0API \u771f\u6b63\u6267\u884c\u5bf9\u5b58\u50a8\u7684\u5199\u5165\u3002 \u8fd9\u4e9b\u914d\u7f6e\u53ef\u5bf9\u6587\u4ef6\u5927\u5c0f\u3001\u538b\u7f29\uff08compression\uff09\u3001\u5e76\u884c\u5ea6\u3001\u538b\u7f29\uff08compaction\uff09\u3001\u5199\u5165\u6a21\u5f0f\u3001\u6e05\u7406\u7b49\u5e95\u5c42\u65b9\u9762\u8fdb\u884c\u5b8c\u5168\u63a7\u5236\u3002 \u5c3d\u7ba1Hudi\u63d0\u4f9b\u4e86\u5408\u7406\u7684\u9ed8\u8ba4\u8bbe\u7f6e\uff0c\u4f46\u5728\u4e0d\u540c\u60c5\u5f62\u4e0b\uff0c\u53ef\u80fd\u9700\u8981\u5bf9\u8fd9\u4e9b\u914d\u7f6e\u8fdb\u884c\u8c03\u6574\u4ee5\u9488\u5bf9\u7279\u5b9a\u7684\u5de5\u4f5c\u8d1f\u8f7d\u8fdb\u884c\u4f18\u5316\u3002<\/li>\n<li>RecordPayload \u914d\u7f6e\u00a0: \u8fd9\u662fHudi\u63d0\u4f9b\u7684\u6700\u5e95\u5c42\u7684\u5b9a\u5236\u3002 RecordPayload\u5b9a\u4e49\u4e86\u5982\u4f55\u6839\u636e\u4f20\u5165\u7684\u65b0\u8bb0\u5f55\u548c\u5b58\u50a8\u7684\u65e7\u8bb0\u5f55\u6765\u4ea7\u751f\u65b0\u503c\u4ee5\u8fdb\u884c\u63d2\u5165\u66f4\u65b0\u3002 Hudi\u63d0\u4f9b\u4e86\u8bf8\u5982OverwriteWithLatestAvroPayload\u7684\u9ed8\u8ba4\u5b9e\u73b0\uff0c\u8be5\u5b9e\u73b0\u4ec5\u4f7f\u7528\u6700\u65b0\u6216\u6700\u540e\u5199\u5165\u7684\u8bb0\u5f55\u6765\u66f4\u65b0\u5b58\u50a8\u3002 \u5728\u6570\u636e\u6e90\u548cWriteClient\u7ea7\u522b\uff0c\u90fd\u53ef\u4ee5\u5c06\u5176\u91cd\u5199\u4e3a\u6269\u5c55HoodieRecordPayload\u7c7b\u7684\u81ea\u5b9a\u4e49\u7c7b\u3002<\/li>\n<\/ul>\n<h2 id=\\\"\u4e0e\u4e91\u5b58\u50a8\u8fde\u63a5\\\">\u4e0e\u4e91\u5b58\u50a8\u8fde\u63a5<\/h2>\n<p>\u65e0\u8bba\u4f7f\u7528RDD\/WriteClient API\u8fd8\u662f\u6570\u636e\u6e90\uff0c\u4ee5\u4e0b\u4fe1\u606f\u90fd\u6709\u52a9\u4e8e\u914d\u7f6e\u5bf9\u4e91\u5b58\u50a8\u7684\u8bbf\u95ee\u3002<\/p>\n<ul>\n<li>AWS S3<br \/>\nS3\u548cHudi\u534f\u540c\u5de5\u4f5c\u6240\u9700\u7684\u914d\u7f6e\u3002<\/li>\n<li>Google Cloud Storage<br \/>\nGCS\u548cHudi\u534f\u540c\u5de5\u4f5c\u6240\u9700\u7684\u914d\u7f6e\u3002<\/li>\n<li>Alibaba Cloud OSS<br \/>\n\u963f\u91cc\u4e91\u548cHudi\u534f\u540c\u5de5\u4f5c\u6240\u9700\u7684\u914d\u7f6e\u3002<\/li>\n<li>Microsoft Azure<br \/>\nAzure\u548cHudi\u534f\u540c\u5de5\u4f5c\u6240\u9700\u7684\u914d\u7f6e\u3002<\/li>\n<\/ul>\n<h2 id=\\\"spark-datasource\\\">Spark\u6570\u636e\u6e90\u914d\u7f6e<\/h2>\n<p>\u53ef\u4ee5\u901a\u8fc7\u5c06\u4ee5\u4e0b\u9009\u9879\u4f20\u9012\u5230option(k,v)\u65b9\u6cd5\u4e2d\u6765\u914d\u7f6e\u4f7f\u7528\u6570\u636e\u6e90\u7684Spark\u4f5c\u4e1a\u3002 \u5b9e\u9645\u7684\u6570\u636e\u6e90\u7ea7\u522b\u914d\u7f6e\u5728\u4e0b\u9762\u5217\u51fa\u3002<\/p>\n<h3 id=\\\"\u5199\u9009\u9879\\\">\u5199\u9009\u9879<\/h3>\n<p>\u53e6\u5916\uff0c\u60a8\u53ef\u4ee5\u4f7f\u7528options()\u6216option(k,v)\u65b9\u6cd5\u76f4\u63a5\u4f20\u9012\u4efb\u4f55WriteClient\u7ea7\u522b\u7684\u914d\u7f6e\u3002<\/p>\n<div>\n<div>\n<pre>inputDF.write()\n.format(\\\"org.apache.hudi\\\")\n.options(clientOpts) \/\/ \u4efb\u4f55Hudi\u5ba2\u6237\u7aef\u9009\u9879\u90fd\u53ef\u4ee5\u4f20\u5165\n.option(DataSourceWriteOptions.RECORDKEY_FIELD_OPT_KEY(), \\\"_row_key\\\")\n.option(DataSourceWriteOptions.PARTITIONPATH_FIELD_OPT_KEY(), \\\"partition\\\")\n.option(DataSourceWriteOptions.PRECOMBINE_FIELD_OPT_KEY(), \\\"timestamp\\\")\n.option(HoodieWriteConfig.TABLE_NAME, tableName)\n.mode(SaveMode.Append)\n.save(basePath);\n<\/pre>\n<\/div>\n<\/div>\n<p>\u7528\u4e8e\u901a\u8fc7write.format.option(&#8230;)\u5199\u5165\u6570\u636e\u96c6\u7684\u9009\u9879<\/p>\n<h4 id=\\\"TABLE_NAME_OPT_KEY\\\">TABLE_NAME_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.table.name\u00a0[\u5fc5\u987b]<br \/>\nHive\u8868\u540d\uff0c\u7528\u4e8e\u5c06\u6570\u636e\u96c6\u6ce8\u518c\u5230\u5176\u4e2d\u3002<\/p>\n<h4 id=\\\"OPERATION_OPT_KEY\\\">OPERATION_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.operation, \u9ed8\u8ba4\u503c\uff1aupsert<br \/>\n\u662f\u5426\u4e3a\u5199\u64cd\u4f5c\u8fdb\u884c\u63d2\u5165\u66f4\u65b0\u3001\u63d2\u5165\u6216\u6279\u91cf\u63d2\u5165\u3002\u4f7f\u7528bulkinsert\u5c06\u65b0\u6570\u636e\u52a0\u8f7d\u5230\u8868\u4e2d\uff0c\u4e4b\u540e\u4f7f\u7528upsert\u6216insert\u3002 \u6279\u91cf\u63d2\u5165\u4f7f\u7528\u57fa\u4e8e\u78c1\u76d8\u7684\u5199\u5165\u8def\u5f84\u6765\u6269\u5c55\u4ee5\u52a0\u8f7d\u5927\u91cf\u8f93\u5165\uff0c\u800c\u65e0\u9700\u5bf9\u5176\u8fdb\u884c\u7f13\u5b58\u3002<\/p>\n<h4 id=\\\"STORAGE_TYPE_OPT_KEY\\\">STORAGE_TYPE_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.storage.type, \u9ed8\u8ba4\u503c\uff1aCOPY_ON_WRITE<br \/>\n\u6b64\u5199\u5165\u7684\u57fa\u7840\u6570\u636e\u7684\u5b58\u50a8\u7c7b\u578b\u3002\u4e24\u6b21\u5199\u5165\u4e4b\u95f4\u4e0d\u80fd\u6539\u53d8\u3002<\/p>\n<h4 id=\\\"PRECOMBINE_FIELD_OPT_KEY\\\">PRECOMBINE_FIELD_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.precombine.field, \u9ed8\u8ba4\u503c\uff1ats<br \/>\n\u5b9e\u9645\u5199\u5165\u4e4b\u524d\u5728preCombining\u4e2d\u4f7f\u7528\u7684\u5b57\u6bb5\u3002 \u5f53\u4e24\u4e2a\u8bb0\u5f55\u5177\u6709\u76f8\u540c\u7684\u952e\u503c\u65f6\uff0c\u6211\u4eec\u5c06\u4f7f\u7528Object.compareTo(..)\u4eceprecombine\u5b57\u6bb5\u4e2d\u9009\u62e9\u4e00\u4e2a\u503c\u6700\u5927\u7684\u8bb0\u5f55\u3002<\/p>\n<h4 id=\\\"PAYLOAD_CLASS_OPT_KEY\\\">PAYLOAD_CLASS_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.payload.class, \u9ed8\u8ba4\u503c\uff1aorg.apache.hudi.OverwriteWithLatestAvroPayload<br \/>\n\u4f7f\u7528\u7684\u6709\u6548\u8f7d\u8377\u7c7b\u3002\u5982\u679c\u60a8\u60f3\u5728\u63d2\u5165\u66f4\u65b0\u6216\u63d2\u5165\u65f6\u4f7f\u7528\u81ea\u5df1\u7684\u5408\u5e76\u903b\u8f91\uff0c\u8bf7\u91cd\u5199\u6b64\u65b9\u6cd5\u3002 \u8fd9\u5c06\u4f7f\u5f97PRECOMBINE_FIELD_OPT_VAL\u8bbe\u7f6e\u7684\u4efb\u4f55\u503c\u65e0\u6548<\/p>\n<h4 id=\\\"RECORDKEY_FIELD_OPT_KEY\\\">RECORDKEY_FIELD_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.recordkey.field, \u9ed8\u8ba4\u503c\uff1auuid<br \/>\n\u8bb0\u5f55\u952e\u5b57\u6bb5\u3002\u7528\u4f5cHoodieKey\u4e2drecordKey\u90e8\u5206\u7684\u503c\u3002 \u5b9e\u9645\u503c\u5c06\u901a\u8fc7\u5728\u5b57\u6bb5\u503c\u4e0a\u8c03\u7528.toString()\u6765\u83b7\u5f97\u3002\u53ef\u4ee5\u4f7f\u7528\u70b9\u7b26\u53f7\u6307\u5b9a\u5d4c\u5957\u5b57\u6bb5\uff0c\u4f8b\u5982\uff1aa.b.c<\/p>\n<h4 id=\\\"PARTITIONPATH_FIELD_OPT_KEY\\\">PARTITIONPATH_FIELD_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.partitionpath.field, \u9ed8\u8ba4\u503c\uff1apartitionpath<br \/>\n\u5206\u533a\u8def\u5f84\u5b57\u6bb5\u3002\u7528\u4f5cHoodieKey\u4e2dpartitionPath\u90e8\u5206\u7684\u503c\u3002 \u901a\u8fc7\u8c03\u7528.toString()\u83b7\u5f97\u5b9e\u9645\u7684\u503c<\/p>\n<h4 id=\\\"KEYGENERATOR_CLASS_OPT_KEY\\\">KEYGENERATOR_CLASS_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.keygenerator.class, \u9ed8\u8ba4\u503c\uff1aorg.apache.hudi.SimpleKeyGenerator<br \/>\n\u952e\u751f\u6210\u5668\u7c7b\uff0c\u5b9e\u73b0\u4ece\u8f93\u5165\u7684Row\u5bf9\u8c61\u4e2d\u63d0\u53d6\u952e<\/p>\n<h4 id=\\\"COMMIT_METADATA_KEYPREFIX_OPT_KEY\\\">COMMIT_METADATA_KEYPREFIX_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.commitmeta.key.prefix, \u9ed8\u8ba4\u503c\uff1a_<br \/>\n\u4ee5\u8be5\u524d\u7f00\u5f00\u5934\u7684\u9009\u9879\u952e\u4f1a\u81ea\u52a8\u6dfb\u52a0\u5230\u63d0\u4ea4\/\u589e\u91cf\u63d0\u4ea4\u7684\u5143\u6570\u636e\u4e2d\u3002 \u8fd9\u5bf9\u4e8e\u4e0ehudi\u65f6\u95f4\u8f74\u4e00\u81f4\u7684\u65b9\u5f0f\u5b58\u50a8\u68c0\u67e5\u70b9\u4fe1\u606f\u5f88\u6709\u7528<\/p>\n<h4 id=\\\"INSERT_DROP_DUPS_OPT_KEY\\\">INSERT_DROP_DUPS_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.write.insert.drop.duplicates, \u9ed8\u8ba4\u503c\uff1afalse<br \/>\n\u5982\u679c\u8bbe\u7f6e\u4e3atrue\uff0c\u5219\u5728\u63d2\u5165\u64cd\u4f5c\u671f\u95f4\u4ece\u4f20\u5165DataFrame\u4e2d\u8fc7\u6ee4\u6389\u6240\u6709\u91cd\u590d\u8bb0\u5f55\u3002<\/p>\n<h4 id=\\\"HIVE_SYNC_ENABLED_OPT_KEY\\\">HIVE_SYNC_ENABLED_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.enable, \u9ed8\u8ba4\u503c\uff1afalse<br \/>\n\u8bbe\u7f6e\u4e3atrue\u65f6\uff0c\u5c06\u6570\u636e\u96c6\u6ce8\u518c\u5e76\u540c\u6b65\u5230Apache Hive Metastore<\/p>\n<h4 id=\\\"HIVE_DATABASE_OPT_KEY\\\">HIVE_DATABASE_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.database, \u9ed8\u8ba4\u503c\uff1adefault<br \/>\n\u8981\u540c\u6b65\u5230\u7684\u6570\u636e\u5e93<\/p>\n<h4 id=\\\"HIVE_TABLE_OPT_KEY\\\">HIVE_TABLE_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.table, [Required]<br \/>\n\u8981\u540c\u6b65\u5230\u7684\u8868<\/p>\n<h4 id=\\\"HIVE_USER_OPT_KEY\\\">HIVE_USER_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.username, \u9ed8\u8ba4\u503c\uff1ahive<br \/>\n\u8981\u4f7f\u7528\u7684Hive\u7528\u6237\u540d<\/p>\n<h4 id=\\\"HIVE_PASS_OPT_KEY\\\">HIVE_PASS_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.password, \u9ed8\u8ba4\u503c\uff1ahive<br \/>\n\u8981\u4f7f\u7528\u7684Hive\u5bc6\u7801<\/p>\n<h4 id=\\\"HIVE_URL_OPT_KEY\\\">HIVE_URL_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.jdbcurl, \u9ed8\u8ba4\u503c\uff1ajdbc:hive2:\/\/localhost:10000<br \/>\nHive metastore url<\/p>\n<h4 id=\\\"HIVE_PARTITION_FIELDS_OPT_KEY\\\">HIVE_PARTITION_FIELDS_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.partition_fields, \u9ed8\u8ba4\u503c\uff1a<br \/>\n\u6570\u636e\u96c6\u4e2d\u7528\u4e8e\u786e\u5b9aHive\u5206\u533a\u7684\u5b57\u6bb5\u3002<\/p>\n<h4 id=\\\"HIVE_PARTITION_EXTRACTOR_CLASS_OPT_KEY\\\">HIVE_PARTITION_EXTRACTOR_CLASS_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.partition_extractor_class, \u9ed8\u8ba4\u503c\uff1aorg.apache.hudi.hive.SlashEncodedDayPartitionValueExtractor<br \/>\n\u7528\u4e8e\u5c06\u5206\u533a\u5b57\u6bb5\u503c\u63d0\u53d6\u5230Hive\u5206\u533a\u5217\u4e2d\u7684\u7c7b\u3002<\/p>\n<h4 id=\\\"HIVE_ASSUME_DATE_PARTITION_OPT_KEY\\\">HIVE_ASSUME_DATE_PARTITION_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.hive_sync.assume_date_partitioning, \u9ed8\u8ba4\u503c\uff1afalse<br \/>\n\u5047\u8bbe\u5206\u533a\u683c\u5f0f\u662fyyyy\/mm\/dd<\/p>\n<h3 id=\\\"\u8bfb\u9009\u9879\\\">\u8bfb\u9009\u9879<\/h3>\n<p>\u7528\u4e8e\u901a\u8fc7read.format.option(&#8230;)\u8bfb\u53d6\u6570\u636e\u96c6\u7684\u9009\u9879<\/p>\n<h4 id=\\\"VIEW_TYPE_OPT_KEY\\\">VIEW_TYPE_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.view.type, \u9ed8\u8ba4\u503c\uff1aread_optimized<br \/>\n\u662f\u5426\u9700\u8981\u4ee5\u67d0\u79cd\u6a21\u5f0f\u8bfb\u53d6\u6570\u636e\uff0c\u589e\u91cf\u6a21\u5f0f\uff08\u81eaInstantTime\u4ee5\u6765\u7684\u65b0\u6570\u636e\uff09 \uff08\u6216\uff09\u8bfb\u4f18\u5316\u6a21\u5f0f\uff08\u57fa\u4e8e\u5217\u6570\u636e\u83b7\u53d6\u6700\u65b0\u89c6\u56fe\uff09 \uff08\u6216\uff09\u5b9e\u65f6\u6a21\u5f0f\uff08\u57fa\u4e8e\u884c\u548c\u5217\u6570\u636e\u83b7\u53d6\u6700\u65b0\u89c6\u56fe\uff09<\/p>\n<h4 id=\\\"BEGIN_INSTANTTIME_OPT_KEY\\\">BEGIN_INSTANTTIME_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.read.begin.instanttime, [\u5728\u589e\u91cf\u6a21\u5f0f\u4e0b\u5fc5\u987b]<br \/>\n\u5f00\u59cb\u589e\u91cf\u63d0\u53d6\u6570\u636e\u7684\u5373\u65f6\u65f6\u95f4\u3002\u8fd9\u91cc\u7684instanttime\u4e0d\u5fc5\u4e00\u5b9a\u4e0e\u65f6\u95f4\u8f74\u4e0a\u7684\u5373\u65f6\u76f8\u5bf9\u5e94\u3002 \u53d6\u51fa\u4ee5instant_time &gt; BEGIN_INSTANTTIME\u5199\u5165\u7684\u65b0\u6570\u636e\u3002 \u4f8b\u5982\uff1a\u201920170901080000\u2019\u5c06\u83b7\u53d62017\u5e749\u67081\u65e508:00 AM\u4e4b\u540e\u5199\u5165\u7684\u6240\u6709\u65b0\u6570\u636e\u3002<\/p>\n<h4 id=\\\"END_INSTANTTIME_OPT_KEY\\\">END_INSTANTTIME_OPT_KEY<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.datasource.read.end.instanttime, \u9ed8\u8ba4\u503c\uff1a\u6700\u65b0\u5373\u65f6\uff08\u5373\u4ece\u5f00\u59cb\u5373\u65f6\u83b7\u53d6\u6240\u6709\u65b0\u6570\u636e\uff09<br \/>\n\u9650\u5236\u589e\u91cf\u63d0\u53d6\u7684\u6570\u636e\u7684\u5373\u65f6\u65f6\u95f4\u3002\u53d6\u51fa\u4ee5instant_time &lt;= END_INSTANTTIME\u5199\u5165\u7684\u65b0\u6570\u636e\u3002<\/p>\n<h2 id=\\\"writeclient-configs\\\">WriteClient \u914d\u7f6e<\/h2>\n<p>\u76f4\u63a5\u4f7f\u7528RDD\u7ea7\u522bapi\u8fdb\u884c\u7f16\u7a0b\u7684Jobs\u53ef\u4ee5\u6784\u5efa\u4e00\u4e2aHoodieWriteConfig\u5bf9\u8c61\uff0c\u5e76\u5c06\u5176\u4f20\u9012\u7ed9HoodieWriteClient\u6784\u9020\u51fd\u6570\u3002 HoodieWriteConfig\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u6784\u5efa\u5668\u6a21\u5f0f\u6784\u5efa\u3002<\/p>\n<div>\n<div>\n<pre>HoodieWriteConfig cfg = HoodieWriteConfig.newBuilder()\n        .withPath(basePath)\n        .forTable(tableName)\n        .withSchema(schemaStr)\n        .withProps(props) \/\/ \u4ece\u5c5e\u6027\u6587\u4ef6\u4f20\u9012\u539f\u59cbk\u3001v\u5bf9\u3002\n        .withCompactionConfig(HoodieCompactionConfig.newBuilder().withXXX(...).build())\n        .withIndexConfig(HoodieIndexConfig.newBuilder().withXXX(...).build())\n        ...\n        .build();\n<\/pre>\n<\/div>\n<\/div>\n<p>\u4ee5\u4e0b\u5404\u8282\u4ecb\u7ecd\u4e86\u5199\u914d\u7f6e\u7684\u4e0d\u540c\u65b9\u9762\uff0c\u5e76\u89e3\u91ca\u4e86\u6700\u91cd\u8981\u7684\u914d\u7f6e\u53ca\u5176\u5c5e\u6027\u540d\u79f0\u548c\u9ed8\u8ba4\u503c\u3002<\/p>\n<h4 id=\\\"withPath\\\">withPath(hoodie_base_path)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.base.path\u00a0[\u5fc5\u987b]<br \/>\n\u521b\u5efa\u6240\u6709\u6570\u636e\u5206\u533a\u6240\u4f9d\u636e\u7684\u57fa\u672cDFS\u8def\u5f84\u3002 \u59cb\u7ec8\u5728\u524d\u7f00\u4e2d\u660e\u786e\u6307\u660e\u5b58\u50a8\u65b9\u5f0f\uff08\u4f8b\u5982hdfs:\/\/\uff0cs3:\/\/\u7b49\uff09\u3002 Hudi\u5c06\u6709\u5173\u63d0\u4ea4\u3001\u4fdd\u5b58\u70b9\u3001\u6e05\u7406\u5ba1\u6838\u65e5\u5fd7\u7b49\u7684\u6240\u6709\u4e3b\u8981\u5143\u6570\u636e\u5b58\u50a8\u5728\u57fa\u672c\u76ee\u5f55\u4e0b\u7684.hoodie\u76ee\u5f55\u4e2d\u3002<\/p>\n<h4 id=\\\"withSchema\\\">withSchema(schema_str)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.avro.schema\u00a0[\u5fc5\u987b]<br \/>\n\u8fd9\u662f\u6570\u636e\u96c6\u7684\u5f53\u524d\u8bfb\u53d6\u5668\u7684avro\u6a21\u5f0f\uff08schema\uff09\u3002 \u8fd9\u662f\u6574\u4e2a\u6a21\u5f0f\u7684\u5b57\u7b26\u4e32\u3002HoodieWriteClient\u4f7f\u7528\u6b64\u6a21\u5f0f\u4f20\u9012\u5230HoodieRecordPayload\u7684\u5b9e\u73b0\uff0c\u4ee5\u4ece\u6e90\u683c\u5f0f\u8f6c\u6362\u4e3aavro\u8bb0\u5f55\u3002 \u5728\u66f4\u65b0\u8fc7\u7a0b\u4e2d\u91cd\u5199\u8bb0\u5f55\u65f6\u4e5f\u4f7f\u7528\u6b64\u6a21\u5f0f\u3002<\/p>\n<h4 id=\\\"forTable\\\">forTable(table_name)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.table.name\u00a0[\u5fc5\u987b]<br \/>\n\u6570\u636e\u96c6\u7684\u8868\u540d\uff0c\u5c06\u7528\u4e8e\u5728Hive\u4e2d\u6ce8\u518c\u3002\u6bcf\u6b21\u8fd0\u884c\u9700\u8981\u76f8\u540c\u3002<\/p>\n<h4 id=\\\"withBulkInsertParallelism\\\">withBulkInsertParallelism(bulk_insert_parallelism = 1500)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.bulkinsert.shuffle.parallelism<br \/>\n\u6279\u91cf\u63d2\u5165\u65e8\u5728\u7528\u4e8e\u8f83\u5927\u7684\u521d\u59cb\u5bfc\u5165\uff0c\u800c\u6b64\u5904\u7684\u5e76\u884c\u5ea6\u51b3\u5b9a\u4e86\u6570\u636e\u96c6\u4e2d\u6587\u4ef6\u7684\u521d\u59cb\u6570\u91cf\u3002 \u8c03\u6574\u6b64\u503c\u4ee5\u8fbe\u5230\u5728\u521d\u59cb\u5bfc\u5165\u671f\u95f4\u6240\u9700\u7684\u6700\u4f73\u5c3a\u5bf8\u3002<\/p>\n<h4 id=\\\"withParallelism\\\">withParallelism(insert_shuffle_parallelism = 1500, upsert_shuffle_parallelism = 1500)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.insert.shuffle.parallelism,\u00a0hoodie.upsert.shuffle.parallelism<br \/>\n\u6700\u521d\u5bfc\u5165\u6570\u636e\u540e\uff0c\u6b64\u5e76\u884c\u5ea6\u5c06\u63a7\u5236\u7528\u4e8e\u8bfb\u53d6\u8f93\u5165\u8bb0\u5f55\u7684\u521d\u59cb\u5e76\u884c\u5ea6\u3002 \u786e\u4fdd\u6b64\u503c\u8db3\u591f\u9ad8\uff0c\u4f8b\u5982\uff1a1\u4e2a\u5206\u533a\u7528\u4e8e1 GB\u7684\u8f93\u5165\u6570\u636e<\/p>\n<h4 id=\\\"combineInput\\\">combineInput(on_insert = false, on_update=true)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.combine.before.insert,\u00a0hoodie.combine.before.upsert<br \/>\n\u5728DFS\u4e2d\u63d2\u5165\u6216\u66f4\u65b0\u4e4b\u524d\u5148\u7ec4\u5408\u8f93\u5165RDD\u5e76\u5c06\u591a\u4e2a\u90e8\u5206\u8bb0\u5f55\u5408\u5e76\u4e3a\u5355\u4e2a\u8bb0\u5f55\u7684\u6807\u5fd7<\/p>\n<h4 id=\\\"withWriteStatusStorageLevel\\\">withWriteStatusStorageLevel(level = MEMORY_AND_DISK_SER)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.write.status.storage.level<br \/>\nHoodieWriteClient.insert\u548cHoodieWriteClient.upsert\u8fd4\u56de\u4e00\u4e2a\u6301\u4e45\u7684RDD[WriteStatus]\uff0c \u8fd9\u662f\u56e0\u4e3a\u5ba2\u6237\u7aef\u53ef\u4ee5\u9009\u62e9\u68c0\u67e5WriteStatus\u5e76\u6839\u636e\u5931\u8d25\u9009\u62e9\u662f\u5426\u63d0\u4ea4\u3002\u8fd9\u662f\u6b64RDD\u7684\u5b58\u50a8\u7ea7\u522b\u7684\u914d\u7f6e<\/p>\n<h4 id=\\\"withAutoCommit\\\">withAutoCommit(autoCommit = true)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.auto.commit<br \/>\n\u63d2\u5165\u548c\u63d2\u5165\u66f4\u65b0\u540e\uff0cHoodieWriteClient\u662f\u5426\u5e94\u8be5\u81ea\u52a8\u63d0\u4ea4\u3002 \u5ba2\u6237\u7aef\u53ef\u4ee5\u9009\u62e9\u5173\u95ed\u81ea\u52a8\u63d0\u4ea4\uff0c\u5e76\u5728\u201d\u5b9a\u4e49\u7684\u6210\u529f\u6761\u4ef6\u201d\u4e0b\u63d0\u4ea4<\/p>\n<h4 id=\\\"withAssumeDatePartitioning\\\">withAssumeDatePartitioning(assumeDatePartitioning = false)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.assume.date.partitioning<br \/>\nHoodieWriteClient\u662f\u5426\u5e94\u8be5\u5047\u8bbe\u6570\u636e\u6309\u65e5\u671f\u5212\u5206\uff0c\u5373\u4ece\u57fa\u672c\u8def\u5f84\u5212\u5206\u4e3a\u4e09\u4e2a\u7ea7\u522b\u3002 \u8fd9\u662f\u652f\u6301&lt;0.3.1\u7248\u672c\u521b\u5efa\u7684\u8868\u7684\u4e00\u4e2a\u8865\u4e01\u3002\u6700\u7ec8\u5c06\u88ab\u5220\u9664<\/p>\n<h4 id=\\\"withConsistencyCheckEnabled\\\">withConsistencyCheckEnabled(enabled = false)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.consistency.check.enabled<br \/>\nHoodieWriteClient\u662f\u5426\u5e94\u8be5\u6267\u884c\u5176\u4ed6\u68c0\u67e5\uff0c\u4ee5\u786e\u4fdd\u5199\u5165\u7684\u6587\u4ef6\u5728\u57fa\u7840\u6587\u4ef6\u7cfb\u7edf\/\u5b58\u50a8\u4e0a\u53ef\u5217\u51fa\u3002 \u5c06\u5176\u8bbe\u7f6e\u4e3atrue\u53ef\u4ee5\u89e3\u51b3S3\u7684\u6700\u7ec8\u4e00\u81f4\u6027\u6a21\u578b\uff0c\u5e76\u786e\u4fdd\u4f5c\u4e3a\u63d0\u4ea4\u7684\u4e00\u90e8\u5206\u5199\u5165\u7684\u6240\u6709\u6570\u636e\u5747\u80fd\u51c6\u786e\u5730\u7528\u4e8e\u67e5\u8be2\u3002<\/p>\n<h3 id=\\\"\u7d22\u5f15\u914d\u7f6e\\\">\u7d22\u5f15\u914d\u7f6e<\/h3>\n<p>\u4ee5\u4e0b\u914d\u7f6e\u63a7\u5236\u7d22\u5f15\u884c\u4e3a\uff0c\u8be5\u884c\u4e3a\u5c06\u4f20\u5165\u8bb0\u5f55\u6807\u8bb0\u4e3a\u5bf9\u8f83\u65e7\u8bb0\u5f55\u7684\u63d2\u5165\u6216\u66f4\u65b0\u3002<br \/>\nwithIndexConfig\u00a0(HoodieIndexConfig)<br \/>\n\u53ef\u63d2\u5165\u4ee5\u5177\u6709\u5916\u90e8\u7d22\u5f15\uff08HBase\uff09\u6216\u4f7f\u7528\u5b58\u50a8\u5728Parquet\u6587\u4ef6\u4e2d\u7684\u9ed8\u8ba4\u5e03\u9686\u8fc7\u6ee4\u5668\uff08bloom filter\uff09<\/p>\n<h4 id=\\\"withIndexClass\\\">withIndexClass(indexClass = \u201cx.y.z.UserDefinedIndex\u201d)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.index.class<br \/>\n\u7528\u6237\u81ea\u5b9a\u4e49\u7d22\u5f15\u7684\u5168\u8def\u5f84\u540d\uff0c\u7d22\u5f15\u7c7b\u5fc5\u987b\u4e3aHoodieIndex\u7684\u5b50\u7c7b\uff0c\u5f53\u6307\u5b9a\u8be5\u914d\u7f6e\u65f6\uff0c\u5176\u4f1a\u4f18\u5148\u4e8ehoodie.index.type\u914d\u7f6e<\/p>\n<h4 id=\\\"withIndexType\\\">withIndexType(indexType = BLOOM)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.index.type<br \/>\n\u8981\u4f7f\u7528\u7684\u7d22\u5f15\u7c7b\u578b\u3002\u9ed8\u8ba4\u4e3a\u5e03\u9686\u8fc7\u6ee4\u5668\u3002\u53ef\u80fd\u7684\u9009\u9879\u662f[BLOOM | HBASE | INMEMORY]\u3002 \u5e03\u9686\u8fc7\u6ee4\u5668\u6d88\u9664\u4e86\u5bf9\u5916\u90e8\u7cfb\u7edf\u7684\u4f9d\u8d56\uff0c\u5e76\u5b58\u50a8\u5728Parquet\u6570\u636e\u6587\u4ef6\u7684\u9875\u811a\u4e2d<\/p>\n<h4 id=\\\"bloomFilterNumEntries\\\">bloomFilterNumEntries(numEntries = 60000)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.index.bloom.num_entries<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aBLOOM\u65f6\u9002\u7528\u3002<br \/>\n\u8fd9\u662f\u8981\u5b58\u50a8\u5728\u5e03\u9686\u8fc7\u6ee4\u5668\u4e2d\u7684\u6761\u76ee\u6570\u3002 \u6211\u4eec\u5047\u8bbemaxParquetFileSize\u4e3a128MB\uff0caverageRecordSize\u4e3a1024B\uff0c\u56e0\u6b64\uff0c\u4e00\u4e2a\u6587\u4ef6\u4e2d\u7684\u8bb0\u5f55\u603b\u6570\u7ea6\u4e3a130K\u3002 \u9ed8\u8ba4\u503c\uff0860000\uff09\u5927\u7ea6\u662f\u6b64\u8fd1\u4f3c\u503c\u7684\u4e00\u534a\u3002HUDI-56\u00a0\u63cf\u8ff0\u4e86\u5982\u4f55\u52a8\u6001\u5730\u5bf9\u6b64\u8fdb\u884c\u8ba1\u7b97\u3002 \u8b66\u544a\uff1a\u5c06\u6b64\u503c\u8bbe\u7f6e\u5f97\u592a\u4f4e\uff0c\u5c06\u4ea7\u751f\u5f88\u591a\u8bef\u62a5\uff0c\u5e76\u4e14\u7d22\u5f15\u67e5\u627e\u5c06\u5fc5\u987b\u626b\u63cf\u6bd4\u5176\u6240\u9700\u7684\u66f4\u591a\u7684\u6587\u4ef6\uff1b\u5982\u679c\u5c06\u5176\u8bbe\u7f6e\u5f97\u975e\u5e38\u9ad8\uff0c\u5c06\u7ebf\u6027\u589e\u52a0\u6bcf\u4e2a\u6570\u636e\u6587\u4ef6\u7684\u5927\u5c0f\uff08\u6bcf50000\u4e2a\u6761\u76ee\u5927\u7ea64KB\uff09\u3002<\/p>\n<h4 id=\\\"bloomFilterFPP\\\">bloomFilterFPP(fpp = 0.000000001)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.index.bloom.fpp<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aBLOOM\u65f6\u9002\u7528\u3002<br \/>\n\u6839\u636e\u6761\u76ee\u6570\u5141\u8bb8\u7684\u9519\u8bef\u7387\u3002 \u8fd9\u7528\u4e8e\u8ba1\u7b97\u5e94\u4e3a\u5e03\u9686\u8fc7\u6ee4\u5668\u5206\u914d\u591a\u5c11\u4f4d\u4ee5\u53ca\u54c8\u5e0c\u51fd\u6570\u7684\u6570\u91cf\u3002\u901a\u5e38\u5c06\u6b64\u503c\u8bbe\u7f6e\u5f97\u5f88\u4f4e\uff08\u9ed8\u8ba4\u503c\uff1a0.000000001\uff09\uff0c\u6211\u4eec\u5e0c\u671b\u5728\u78c1\u76d8\u7a7a\u95f4\u4e0a\u8fdb\u884c\u6743\u8861\u4ee5\u964d\u4f4e\u8bef\u62a5\u7387<\/p>\n<h4 id=\\\"bloomIndexPruneByRanges\\\">bloomIndexPruneByRanges(pruneRanges = true)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.bloom.index.prune.by.ranges<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aBLOOM\u65f6\u9002\u7528\u3002<br \/>\n\u4e3atrue\u65f6\uff0c\u4ece\u6587\u4ef6\u6846\u5b9a\u4fe1\u606f\uff0c\u53ef\u4ee5\u52a0\u5feb\u7d22\u5f15\u67e5\u627e\u7684\u901f\u5ea6\u3002 \u5982\u679c\u952e\u5177\u6709\u5355\u8c03\u9012\u589e\u7684\u524d\u7f00\uff0c\u4f8b\u5982\u65f6\u95f4\u6233\uff0c\u5219\u7279\u522b\u6709\u7528\u3002<\/p>\n<h4 id=\\\"bloomIndexUseCaching\\\">bloomIndexUseCaching(useCaching = true)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.bloom.index.use.caching<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aBLOOM\u65f6\u9002\u7528\u3002<br \/>\n\u4e3atrue\u65f6\uff0c\u5c06\u901a\u8fc7\u51cf\u5c11\u7528\u4e8e\u8ba1\u7b97\u5e76\u884c\u5ea6\u6216\u53d7\u5f71\u54cd\u5206\u533a\u7684IO\u6765\u7f13\u5b58\u8f93\u5165\u7684RDD\u4ee5\u52a0\u5feb\u7d22\u5f15\u67e5\u627e<\/p>\n<h4 id=\\\"bloomIndexTreebasedFilter\\\">bloomIndexTreebasedFilter(useTreeFilter = true)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.bloom.index.use.treebased.filter<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aBLOOM\u65f6\u9002\u7528\u3002<br \/>\n\u4e3atrue\u65f6\uff0c\u542f\u7528\u57fa\u4e8e\u95f4\u9694\u6811\u7684\u6587\u4ef6\u8fc7\u6ee4\u4f18\u5316\u3002\u4e0e\u66b4\u529b\u6a21\u5f0f\u76f8\u6bd4\uff0c\u6b64\u6a21\u5f0f\u53ef\u6839\u636e\u952e\u8303\u56f4\u52a0\u5feb\u6587\u4ef6\u8fc7\u6ee4\u901f\u5ea6<\/p>\n<h4 id=\\\"bloomIndexBucketizedChecking\\\">bloomIndexBucketizedChecking(bucketizedChecking = true)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.bloom.index.bucketized.checking<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aBLOOM\u65f6\u9002\u7528\u3002<br \/>\n\u4e3atrue\u65f6\uff0c\u542f\u7528\u4e86\u6876\u5f0f\u5e03\u9686\u8fc7\u6ee4\u3002\u8fd9\u51cf\u5c11\u4e86\u5728\u57fa\u4e8e\u6392\u5e8f\u7684\u5e03\u9686\u7d22\u5f15\u67e5\u627e\u4e2d\u770b\u5230\u7684\u504f\u5dee<\/p>\n<h4 id=\\\"bloomIndexKeysPerBucket\\\">bloomIndexKeysPerBucket(keysPerBucket = 10000000)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.bloom.index.keys.per.bucket<br \/>\n\u4ec5\u5728\u542f\u7528bloomIndexBucketizedChecking\u5e76\u4e14\u7d22\u5f15\u7c7b\u578b\u4e3abloom\u7684\u60c5\u51b5\u4e0b\u9002\u7528\u3002<br \/>\n\u6b64\u914d\u7f6e\u63a7\u5236\u201c\u5b58\u50a8\u6876\u201d\u7684\u5927\u5c0f\uff0c\u8be5\u5927\u5c0f\u53ef\u8ddf\u8e2a\u5bf9\u5355\u4e2a\u6587\u4ef6\u8fdb\u884c\u7684\u8bb0\u5f55\u952e\u68c0\u67e5\u7684\u6b21\u6570\uff0c\u5e76\u4e14\u662f\u5206\u914d\u7ed9\u6267\u884c\u5e03\u9686\u8fc7\u6ee4\u5668\u67e5\u627e\u7684\u6bcf\u4e2a\u5206\u533a\u7684\u5de5\u4f5c\u5355\u4f4d\u3002 \u8f83\u9ad8\u7684\u503c\u5c06\u5206\u644a\u5c06\u5e03\u9686\u8fc7\u6ee4\u5668\u8bfb\u53d6\u5230\u5185\u5b58\u7684\u56fa\u5b9a\u6210\u672c\u3002<\/p>\n<h4 id=\\\"bloomIndexParallelism\\\">bloomIndexParallelism(0)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.bloom.index.parallelism<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aBLOOM\u65f6\u9002\u7528\u3002<br \/>\n\u8fd9\u662f\u7d22\u5f15\u67e5\u627e\u7684\u5e76\u884c\u5ea6\uff0c\u5176\u4e2d\u6d89\u53caSpark Shuffle\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u8fd9\u662f\u6839\u636e\u8f93\u5165\u7684\u5de5\u4f5c\u8d1f\u8f7d\u7279\u5f81\u81ea\u52a8\u8ba1\u7b97\u7684<\/p>\n<h4 id=\\\"hbaseZkQuorum\\\">hbaseZkQuorum(zkString) [\u5fc5\u987b]<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.index.hbase.zkquorum<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aHBASE\u65f6\u9002\u7528\u3002\u8981\u8fde\u63a5\u7684HBase ZK Quorum URL\u3002<\/p>\n<h4 id=\\\"hbaseZkPort\\\">hbaseZkPort(port) [\u5fc5\u987b]<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.index.hbase.zkport<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aHBASE\u65f6\u9002\u7528\u3002\u8981\u8fde\u63a5\u7684HBase ZK Quorum\u7aef\u53e3\u3002<\/p>\n<h4 id=\\\"hbaseTableName\\\">hbaseZkZnodeParent(zkZnodeParent) [\u5fc5\u987b]<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.index.hbase.zknode.path<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aHBASE\u65f6\u9002\u7528\u3002\u8fd9\u662f\u6839znode\uff0c\u5b83\u5c06\u5305\u542bHBase\u521b\u5efa\u53ca\u4f7f\u7528\u7684\u6240\u6709znode\u3002<\/p>\n<h4 id=\\\"hbaseTableName\\\">hbaseTableName(tableName) [\u5fc5\u987b]<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.index.hbase.table<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aHBASE\u65f6\u9002\u7528\u3002HBase\u8868\u540d\u79f0\uff0c\u7528\u4f5c\u7d22\u5f15\u3002Hudi\u5c06row_key\u548c[partition_path, fileID, commitTime]\u6620\u5c04\u5b58\u50a8\u5728\u8868\u4e2d\u3002<\/p>\n<h5 id=\\\"bloomIndexUpdatePartitionPath\\\">bloomIndexUpdatePartitionPath(updatePartitionPath = false)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.bloom.index.update.partition.path<br \/>\n\u4ec5\u5728\u7d22\u5f15\u7c7b\u578b\u4e3aGLOBAL_BLOOM\u65f6\u9002\u7528\u3002<br \/>\n\u4e3atrue\u65f6\uff0c\u5f53\u5bf9\u4e00\u4e2a\u5df2\u6709\u8bb0\u5f55\u6267\u884c\u5305\u542b\u5206\u533a\u8def\u5f84\u7684\u66f4\u65b0\u64cd\u4f5c\u65f6\uff0c\u5c06\u4f1a\u5bfc\u81f4\u628a\u65b0\u8bb0\u5f55\u63d2\u5165\u5230\u65b0\u5206\u533a\uff0c\u800c\u628a\u539f\u6709\u8bb0\u5f55\u4ece\u65e7\u5206\u533a\u91cc\u5220\u9664\u3002\u4e3afalse\u65f6\uff0c\u53ea\u5bf9\u65e7\u5206\u533a\u7684\u539f\u6709\u8bb0\u5f55\u8fdb\u884c\u66f4\u65b0\u3002<\/p>\n<h3 id=\\\"\u5b58\u50a8\u9009\u9879\\\">\u5b58\u50a8\u9009\u9879<\/h3>\n<p>\u63a7\u5236\u6709\u5173\u8c03\u6574parquet\u548c\u65e5\u5fd7\u6587\u4ef6\u5927\u5c0f\u7684\u65b9\u9762\u3002<br \/>\nwithStorageConfig\u00a0(HoodieStorageConfig)<\/p>\n<h4 id=\\\"limitFileSize\\\">limitFileSize (size = 120MB)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.parquet.max.file.size<br \/>\nHudi\u5199\u9636\u6bb5\u751f\u6210\u7684parquet\u6587\u4ef6\u7684\u76ee\u6807\u5927\u5c0f\u3002\u5bf9\u4e8eDFS\uff0c\u8fd9\u9700\u8981\u4e0e\u57fa\u7840\u6587\u4ef6\u7cfb\u7edf\u5757\u5927\u5c0f\u4fdd\u6301\u4e00\u81f4\uff0c\u4ee5\u5b9e\u73b0\u6700\u4f73\u6027\u80fd\u3002<\/p>\n<h4 id=\\\"parquetBlockSize\\\">parquetBlockSize(rowgroupsize = 120MB)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.parquet.block.size<br \/>\nParquet\u884c\u7ec4\u5927\u5c0f\u3002\u6700\u597d\u4e0e\u6587\u4ef6\u5927\u5c0f\u76f8\u540c\uff0c\u4ee5\u4fbf\u5c06\u6587\u4ef6\u4e2d\u7684\u5355\u4e2a\u5217\u8fde\u7eed\u5b58\u50a8\u5728\u78c1\u76d8\u4e0a<\/p>\n<h4 id=\\\"parquetPageSize\\\">parquetPageSize(pagesize = 1MB)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.parquet.page.size<br \/>\nParquet\u9875\u9762\u5927\u5c0f\u3002\u9875\u9762\u662fparquet\u6587\u4ef6\u4e2d\u7684\u8bfb\u53d6\u5355\u4f4d\u3002 \u5728\u4e00\u4e2a\u5757\u5185\uff0c\u9875\u9762\u88ab\u5206\u522b\u538b\u7f29\u3002<\/p>\n<h4 id=\\\"parquetCompressionRatio\\\">parquetCompressionRatio(parquetCompressionRatio = 0.1)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.parquet.compression.ratio<br \/>\n\u5f53Hudi\u5c1d\u8bd5\u8c03\u6574\u65b0parquet\u6587\u4ef6\u7684\u5927\u5c0f\u65f6\uff0c\u9884\u671f\u5bf9parquet\u6570\u636e\u8fdb\u884c\u538b\u7f29\u7684\u6bd4\u4f8b\u3002 \u5982\u679cbulk_insert\u751f\u6210\u7684\u6587\u4ef6\u5c0f\u4e8e\u9884\u671f\u5927\u5c0f\uff0c\u8bf7\u589e\u52a0\u6b64\u503c<\/p>\n<h4 id=\\\"parquetCompressionCodec\\\">parquetCompressionCodec(parquetCompressionCodec = gzip)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.parquet.compression.codec<br \/>\nParquet\u538b\u7f29\u7f16\u89e3\u7801\u65b9\u5f0f\u540d\u79f0\u3002\u9ed8\u8ba4\u503c\u4e3agzip\u3002\u53ef\u80fd\u7684\u9009\u9879\u662f[gzip | snappy | uncompressed | lzo]<\/p>\n<h4 id=\\\"logFileMaxSize\\\">logFileMaxSize(logFileSize = 1GB)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.logfile.max.size<br \/>\nLogFile\u7684\u6700\u5927\u5927\u5c0f\u3002\u8fd9\u662f\u5728\u5c06\u65e5\u5fd7\u6587\u4ef6\u79fb\u5230\u4e0b\u4e00\u4e2a\u7248\u672c\u4e4b\u524d\u5141\u8bb8\u7684\u6700\u5927\u5927\u5c0f\u3002<\/p>\n<h4 id=\\\"logFileDataBlockMaxSize\\\">logFileDataBlockMaxSize(dataBlockSize = 256MB)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.logfile.data.block.max.size<br \/>\nLogFile\u6570\u636e\u5757\u7684\u6700\u5927\u5927\u5c0f\u3002\u8fd9\u662f\u5141\u8bb8\u5c06\u5355\u4e2a\u6570\u636e\u5757\u9644\u52a0\u5230\u65e5\u5fd7\u6587\u4ef6\u7684\u6700\u5927\u5927\u5c0f\u3002 \u8fd9\u6709\u52a9\u4e8e\u786e\u4fdd\u9644\u52a0\u5230\u65e5\u5fd7\u6587\u4ef6\u7684\u6570\u636e\u88ab\u5206\u89e3\u4e3a\u53ef\u8c03\u6574\u5927\u5c0f\u7684\u5757\uff0c\u4ee5\u9632\u6b62\u53d1\u751fOOM\u9519\u8bef\u3002\u6b64\u5927\u5c0f\u5e94\u5927\u4e8eJVM\u5185\u5b58\u3002<\/p>\n<h4 id=\\\"logFileToParquetCompressionRatio\\\">logFileToParquetCompressionRatio(logFileToParquetCompressionRatio = 0.35)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.logfile.to.parquet.compression.ratio<br \/>\n\u968f\u7740\u8bb0\u5f55\u4ece\u65e5\u5fd7\u6587\u4ef6\u79fb\u52a8\u5230parquet\uff0c\u9884\u671f\u4f1a\u8fdb\u884c\u989d\u5916\u538b\u7f29\u7684\u6bd4\u4f8b\u3002 \u7528\u4e8emerge_on_read\u5b58\u50a8\uff0c\u4ee5\u5c06\u63d2\u5165\u5185\u5bb9\u53d1\u9001\u5230\u65e5\u5fd7\u6587\u4ef6\u4e2d\u5e76\u63a7\u5236\u538b\u7f29parquet\u6587\u4ef6\u7684\u5927\u5c0f\u3002<\/p>\n<h4 id=\\\"parquetCompressionCodec\\\">parquetCompressionCodec(parquetCompressionCodec = gzip)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.parquet.compression.codec<br \/>\nParquet\u6587\u4ef6\u7684\u538b\u7f29\u7f16\u89e3\u7801\u65b9\u5f0f<\/p>\n<h3 id=\\\"\u538b\u7f29\u914d\u7f6e\\\">\u538b\u7f29\u914d\u7f6e<\/h3>\n<p>\u538b\u7f29\u914d\u7f6e\u7528\u4e8e\u63a7\u5236\u538b\u7f29\uff08\u5c06\u65e5\u5fd7\u6587\u4ef6\u5408\u5e76\u5230\u65b0\u7684parquet\u57fa\u672c\u6587\u4ef6\u4e2d\uff09\u3001\u6e05\u7406\uff08\u56de\u6536\u8f83\u65e7\u53ca\u672a\u4f7f\u7528\u7684\u6587\u4ef6\u7ec4\uff09\u3002\u00a0withCompactionConfig\u00a0(HoodieCompactionConfig)<\/p>\n<h4 id=\\\"withCleanerPolicy\\\">withCleanerPolicy(policy = KEEP_LATEST_COMMITS)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.cleaner.policy<br \/>\n\u8981\u4f7f\u7528\u7684\u6e05\u7406\u653f\u7b56\u3002Hudi\u5c06\u5220\u9664\u65e7\u7248\u672c\u7684parquet\u6587\u4ef6\u4ee5\u56de\u6536\u7a7a\u95f4\u3002 \u4efb\u4f55\u5f15\u7528\u6b64\u7248\u672c\u6587\u4ef6\u7684\u67e5\u8be2\u548c\u8ba1\u7b97\u90fd\u5c06\u5931\u8d25\u3002\u6700\u597d\u786e\u4fdd\u6570\u636e\u4fdd\u7559\u7684\u65f6\u95f4\u8d85\u8fc7\u6700\u5927\u67e5\u8be2\u6267\u884c\u65f6\u95f4\u3002<\/p>\n<h4 id=\\\"retainCommits\\\">retainCommits(no_of_commits_to_retain = 24)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.cleaner.commits.retained<br \/>\n\u4fdd\u7559\u7684\u63d0\u4ea4\u6570\u3002\u56e0\u6b64\uff0c\u6570\u636e\u5c06\u4fdd\u7559\u4e3anum_of_commits * time_between_commits\uff08\u8ba1\u5212\u7684\uff09\u3002 \u8fd9\u4e5f\u76f4\u63a5\u8f6c\u5316\u4e3a\u60a8\u53ef\u4ee5\u9010\u6b65\u63d0\u53d6\u6b64\u6570\u636e\u96c6\u7684\u6570\u91cf<\/p>\n<h4 id=\\\"archiveCommitsWith\\\">archiveCommitsWith(minCommits = 96, maxCommits = 128)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.keep.min.commits,\u00a0hoodie.keep.max.commits<br \/>\n\u6bcf\u4e2a\u63d0\u4ea4\u90fd\u662f.hoodie\u76ee\u5f55\u4e2d\u7684\u4e00\u4e2a\u5c0f\u6587\u4ef6\u3002\u7531\u4e8eDFS\u901a\u5e38\u4e0d\u652f\u6301\u5927\u91cf\u5c0f\u6587\u4ef6\uff0c\u56e0\u6b64Hudi\u5c06\u8f83\u65e9\u7684\u63d0\u4ea4\u5f52\u6863\u5230\u987a\u5e8f\u65e5\u5fd7\u4e2d\u3002 \u63d0\u4ea4\u901a\u8fc7\u91cd\u547d\u540d\u63d0\u4ea4\u6587\u4ef6\u4ee5\u539f\u5b50\u65b9\u5f0f\u53d1\u5e03\u3002<\/p>\n<h4 id=\\\"withCommitsArchivalBatchSize\\\">withCommitsArchivalBatchSize(batch = 10)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.commits.archival.batch<br \/>\n\u8fd9\u63a7\u5236\u7740\u6279\u91cf\u8bfb\u53d6\u5e76\u4e00\u8d77\u5f52\u6863\u7684\u63d0\u4ea4\u5373\u65f6\u7684\u6570\u91cf\u3002<\/p>\n<h4 id=\\\"compactionSmallFileSize\\\">compactionSmallFileSize(size = 0)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.parquet.small.file.limit<br \/>\n\u8be5\u503c\u5e94\u5c0f\u4e8emaxFileSize\uff0c\u5982\u679c\u5c06\u5176\u8bbe\u7f6e\u4e3a0\uff0c\u4f1a\u5173\u95ed\u6b64\u529f\u80fd\u3002 \u7531\u4e8e\u6279\u5904\u7406\u4e2d\u5206\u533a\u4e2d\u63d2\u5165\u8bb0\u5f55\u7684\u6570\u91cf\u4f17\u591a\uff0c\u603b\u4f1a\u51fa\u73b0\u5c0f\u6587\u4ef6\u3002 Hudi\u63d0\u4f9b\u4e86\u4e00\u4e2a\u9009\u9879\uff0c\u53ef\u4ee5\u901a\u8fc7\u5c06\u5bf9\u8be5\u5206\u533a\u4e2d\u7684\u63d2\u5165\u4f5c\u4e3a\u5bf9\u73b0\u6709\u5c0f\u6587\u4ef6\u7684\u66f4\u65b0\u6765\u89e3\u51b3\u5c0f\u6587\u4ef6\u7684\u95ee\u9898\u3002 \u6b64\u5904\u7684\u5927\u5c0f\u662f\u88ab\u89c6\u4e3a\u201c\u5c0f\u6587\u4ef6\u5927\u5c0f\u201d\u7684\u6700\u5c0f\u6587\u4ef6\u5927\u5c0f\u3002<\/p>\n<h4 id=\\\"insertSplitSize\\\">insertSplitSize(size = 500000)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.copyonwrite.insert.split.size<br \/>\n\u63d2\u5165\u5199\u5165\u5e76\u884c\u5ea6\u3002\u4e3a\u5355\u4e2a\u5206\u533a\u7684\u603b\u5171\u63d2\u5165\u6b21\u6570\u3002 \u5199\u51fa100MB\u7684\u6587\u4ef6\uff0c\u81f3\u5c111kb\u5927\u5c0f\u7684\u8bb0\u5f55\uff0c\u610f\u5473\u7740\u6bcf\u4e2a\u6587\u4ef6\u6709100K\u8bb0\u5f55\u3002\u9ed8\u8ba4\u503c\u662f\u8d85\u989d\u914d\u7f6e\u4e3a500K\u3002 \u4e3a\u4e86\u6539\u5584\u63d2\u5165\u5ef6\u8fdf\uff0c\u8bf7\u5bf9\u5176\u8fdb\u884c\u8c03\u6574\u4ee5\u5339\u914d\u5355\u4e2a\u6587\u4ef6\u4e2d\u7684\u8bb0\u5f55\u6570\u3002 \u5c06\u6b64\u503c\u8bbe\u7f6e\u4e3a\u8f83\u5c0f\u7684\u503c\u5c06\u5bfc\u81f4\u6587\u4ef6\u53d8\u5c0f\uff08\u5c24\u5176\u662f\u5f53compactionSmallFileSize\u4e3a0\u65f6\uff09<\/p>\n<h4 id=\\\"autoTuneInsertSplits\\\">autoTuneInsertSplits(true)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.copyonwrite.insert.auto.split<br \/>\nHudi\u662f\u5426\u5e94\u8be5\u57fa\u4e8e\u6700\u540e24\u4e2a\u63d0\u4ea4\u7684\u5143\u6570\u636e\u52a8\u6001\u8ba1\u7b97insertSplitSize\u3002\u9ed8\u8ba4\u5173\u95ed\u3002<\/p>\n<h4 id=\\\"approxRecordSize\\\">approxRecordSize(size = 1024)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.copyonwrite.record.size.estimate<br \/>\n\u5e73\u5747\u8bb0\u5f55\u5927\u5c0f\u3002\u5982\u679c\u6307\u5b9a\uff0chudi\u5c06\u4f7f\u7528\u5b83\uff0c\u5e76\u4e14\u4e0d\u4f1a\u57fa\u4e8e\u6700\u540e24\u4e2a\u63d0\u4ea4\u7684\u5143\u6570\u636e\u52a8\u6001\u5730\u8ba1\u7b97\u3002 \u6ca1\u6709\u9ed8\u8ba4\u503c\u8bbe\u7f6e\u3002\u8fd9\u5bf9\u4e8e\u8ba1\u7b97\u63d2\u5165\u5e76\u884c\u5ea6\u4ee5\u53ca\u5c06\u63d2\u5165\u6253\u5305\u5230\u5c0f\u6587\u4ef6\u4e2d\u81f3\u5173\u91cd\u8981\u3002\u5982\u4e0a\u6240\u8ff0\u3002<\/p>\n<h4 id=\\\"withInlineCompaction\\\">withInlineCompaction(inlineCompaction = false)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.compact.inline<br \/>\n\u5f53\u8bbe\u7f6e\u4e3atrue\u65f6\uff0c\u7d27\u63a5\u5728\u63d2\u5165\u6216\u63d2\u5165\u66f4\u65b0\u6216\u6279\u91cf\u63d2\u5165\u7684\u63d0\u4ea4\u6216\u589e\u91cf\u63d0\u4ea4\u64cd\u4f5c\u4e4b\u540e\u7531\u6444\u53d6\u672c\u8eab\u89e6\u53d1\u538b\u7f29<\/p>\n<h4 id=\\\"withMaxNumDeltaCommitsBeforeCompaction\\\">withMaxNumDeltaCommitsBeforeCompaction(maxNumDeltaCommitsBeforeCompaction = 10)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.compact.inline.max.delta.commits<br \/>\n\u89e6\u53d1\u5185\u8054\u538b\u7f29\u4e4b\u524d\u8981\u4fdd\u7559\u7684\u6700\u5927\u589e\u91cf\u63d0\u4ea4\u6570<\/p>\n<h4 id=\\\"withCompactionLazyBlockReadEnabled\\\">withCompactionLazyBlockReadEnabled(true)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.compaction.lazy.block.read<br \/>\n\u5f53CompactedLogScanner\u5408\u5e76\u6240\u6709\u65e5\u5fd7\u6587\u4ef6\u65f6\uff0c\u6b64\u914d\u7f6e\u6709\u52a9\u4e8e\u9009\u62e9\u662f\u5426\u5e94\u5ef6\u8fdf\u8bfb\u53d6\u65e5\u5fd7\u5757\u3002 \u9009\u62e9true\u4ee5\u4f7f\u7528I\/O\u5bc6\u96c6\u578b\u5ef6\u8fdf\u5757\u8bfb\u53d6\uff08\u4f4e\u5185\u5b58\u4f7f\u7528\uff09\uff0c\u6216\u8005\u4e3afalse\u6765\u4f7f\u7528\u5185\u5b58\u5bc6\u96c6\u578b\u7acb\u5373\u5757\u8bfb\u53d6\uff08\u9ad8\u5185\u5b58\u4f7f\u7528\uff09<\/p>\n<h4 id=\\\"withCompactionReverseLogReadEnabled\\\">withCompactionReverseLogReadEnabled(false)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.compaction.reverse.log.read<br \/>\nHoodieLogFormatReader\u4f1a\u4ecepos=0\u5230pos=file_length\u5411\u524d\u8bfb\u53d6\u65e5\u5fd7\u6587\u4ef6\u3002 \u5982\u679c\u6b64\u914d\u7f6e\u8bbe\u7f6e\u4e3atrue\uff0c\u5219Reader\u4f1a\u4ecepos=file_length\u5230pos=0\u53cd\u5411\u8bfb\u53d6\u65e5\u5fd7\u6587\u4ef6<\/p>\n<h4 id=\\\"withCleanerParallelism\\\">withCleanerParallelism(cleanerParallelism = 200)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.cleaner.parallelism<br \/>\n\u5982\u679c\u6e05\u7406\u53d8\u6162\uff0c\u8bf7\u589e\u52a0\u6b64\u503c\u3002<\/p>\n<h4 id=\\\"withCompactionStrategy\\\">withCompactionStrategy(compactionStrategy = org.apache.hudi.io.compact.strategy.LogFileSizeBasedCompactionStrategy)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.compaction.strategy<br \/>\n\u7528\u6765\u51b3\u5b9a\u5728\u6bcf\u6b21\u538b\u7f29\u8fd0\u884c\u671f\u95f4\u9009\u62e9\u8981\u538b\u7f29\u7684\u6587\u4ef6\u7ec4\u7684\u538b\u7f29\u7b56\u7565\u3002 \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0cHudi\u9009\u62e9\u5177\u6709\u7d2f\u79ef\u6700\u591a\u672a\u5408\u5e76\u6570\u636e\u7684\u65e5\u5fd7\u6587\u4ef6<\/p>\n<h4 id=\\\"withTargetIOPerCompactionInMB\\\">withTargetIOPerCompactionInMB(targetIOPerCompactionInMB = 500000)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.compaction.target.io<br \/>\nLogFileSizeBasedCompactionStrategy\u7684\u538b\u7f29\u8fd0\u884c\u671f\u95f4\u8981\u82b1\u8d39\u7684MB\u91cf\u3002\u5f53\u538b\u7f29\u4ee5\u5185\u8054\u6a21\u5f0f\u8fd0\u884c\u65f6\uff0c\u6b64\u503c\u6709\u52a9\u4e8e\u9650\u5236\u6444\u53d6\u5ef6\u8fdf\u3002<\/p>\n<h4 id=\\\"withTargetPartitionsPerDayBasedCompaction\\\">withTargetPartitionsPerDayBasedCompaction(targetPartitionsPerCompaction = 10)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.compaction.daybased.target<br \/>\n\u7531org.apache.hudi.io.compact.strategy.DayBasedCompactionStrategy\u4f7f\u7528\uff0c\u8868\u793a\u5728\u538b\u7f29\u8fd0\u884c\u671f\u95f4\u8981\u538b\u7f29\u7684\u6700\u65b0\u5206\u533a\u6570\u3002<\/p>\n<h4 id=\\\"payloadClassName\\\">withPayloadClass(payloadClassName = org.apache.hudi.common.model.HoodieAvroPayload)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.compaction.payload.class<br \/>\n\u8fd9\u9700\u8981\u4e0e\u63d2\u5165\/\u63d2\u5165\u66f4\u65b0\u8fc7\u7a0b\u4e2d\u4f7f\u7528\u7684\u7c7b\u76f8\u540c\u3002 \u5c31\u50cf\u5199\u5165\u4e00\u6837\uff0c\u538b\u7f29\u4e5f\u4f7f\u7528\u8bb0\u5f55\u6709\u6548\u8d1f\u8f7d\u7c7b\u5c06\u65e5\u5fd7\u4e2d\u7684\u8bb0\u5f55\u5f7c\u6b64\u5408\u5e76\uff0c\u518d\u6b21\u4e0e\u57fa\u672c\u6587\u4ef6\u5408\u5e76\uff0c\u5e76\u751f\u6210\u538b\u7f29\u540e\u8981\u5199\u5165\u7684\u6700\u7ec8\u8bb0\u5f55\u3002<\/p>\n<h3 id=\\\"\u6307\u6807\u914d\u7f6e\\\">\u6307\u6807\u914d\u7f6e<\/h3>\n<p>\u914d\u7f6eHudi\u6307\u6807\u62a5\u544a\u3002\u00a0withMetricsConfig\u00a0(HoodieMetricsConfig)<br \/>\nHudi\u4f1a\u53d1\u5e03\u6709\u5173\u6bcf\u6b21\u63d0\u4ea4\u3001\u6e05\u7406\u3001\u56de\u6eda\u7b49\u7684\u6307\u6807\u3002<\/p>\n<h4 id=\\\"graphite\\\">GRAPHITEPermalink<\/h4>\n<h5 id=\\\"on\\\">on(metricsOn = true)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.on<br \/>\n\u6253\u5f00\u6216\u5173\u95ed\u53d1\u9001\u6307\u6807\u3002\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u5904\u4e8e\u542f\u7528\u72b6\u6001\u3002<\/p>\n<h5 id=\\\"withReporterType\\\">withReporterType(reporterType = GRAPHITE)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.reporter.type<br \/>\n\u6307\u6807\u62a5\u544a\u8005\u7684\u7c7b\u578b\u3002\u9ed8\u8ba4\u4f7f\u7528graphite\u3002<\/p>\n<h5 id=\\\"toGraphiteHost\\\">toGraphiteHost(host = localhost)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.graphite.host<br \/>\n\u8981\u8fde\u63a5\u7684graphite\u4e3b\u673a<\/p>\n<h5 id=\\\"onGraphitePort\\\">onGraphitePort(port = 4756)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.graphite.port<br \/>\n\u8981\u8fde\u63a5\u7684graphite\u7aef\u53e3<\/p>\n<h5 id=\\\"usePrefix\\\">usePrefix(prefix = \u201c\u201d)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.graphite.metric.prefix<br \/>\n\u9002\u7528\u4e8e\u6240\u6709\u6307\u6807\u7684\u6807\u51c6\u524d\u7f00\u3002\u8fd9\u6709\u52a9\u4e8e\u6dfb\u52a0\u5982\u6570\u636e\u4e2d\u5fc3\u3001\u73af\u5883\u7b49\u4fe1\u606f<\/p>\n<h4 id=\\\"jmx\\\">JMXPermalink<\/h4>\n<h5 id=\\\"on\\\">on(metricsOn = true)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.on<br \/>\n\u6253\u5f00\u6216\u5173\u95ed\u53d1\u9001\u6307\u6807\u3002\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u5904\u4e8e\u542f\u7528\u72b6\u6001\u3002<\/p>\n<h5 id=\\\"withReporterType\\\">withReporterType(reporterType = JMX)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.reporter.type<br \/>\n\u6307\u6807\u62a5\u544a\u8005\u7684\u7c7b\u578b\u3002<\/p>\n<h5 id=\\\"toJmxHost\\\">toJmxHost(host = localhost)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.jmx.host<br \/>\n\u8981\u8fde\u63a5\u7684Jmx\u4e3b\u673a<\/p>\n<h5 id=\\\"onJmxPort\\\">onJmxPort(port = 1000-5000)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.graphite.port<br \/>\n\u8981\u8fde\u63a5\u7684Jmx\u7aef\u53e3<\/p>\n<h5 id=\\\"usePrefix\\\">usePrefix(prefix = \u201c\u201d)<\/h5>\n<p>\u5c5e\u6027\uff1ahoodie.metrics.jmx.metric.prefix<br \/>\n\u9002\u7528\u4e8e\u6240\u6709\u6307\u6807\u7684\u6807\u51c6\u524d\u7f00\u3002\u8fd9\u6709\u52a9\u4e8e\u6dfb\u52a0\u5982\u6570\u636e\u4e2d\u5fc3\u3001\u73af\u5883\u7b49\u4fe1\u606f<\/p>\n<h3 id=\\\"\u5185\u5b58\u914d\u7f6e\\\">\u5185\u5b58\u914d\u7f6e<\/h3>\n<p>\u63a7\u5236\u7531Hudi\u5185\u90e8\u6267\u884c\u7684\u538b\u7f29\u548c\u5408\u5e76\u7684\u5185\u5b58\u4f7f\u7528\u60c5\u51b5\u00a0withMemoryConfig\u00a0(HoodieMemoryConfig)<br \/>\n\u5185\u5b58\u76f8\u5173\u914d\u7f6e<\/p>\n<h4 id=\\\"withMaxMemoryFractionPerPartitionMerge\\\">withMaxMemoryFractionPerPartitionMerge(maxMemoryFractionPerPartitionMerge = 0.6)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.memory.merge.fraction<br \/>\n\u8be5\u6bd4\u4f8b\u4e58\u4ee5\u7528\u6237\u5185\u5b58\u6bd4\u4f8b\uff081-spark.memory.fraction\uff09\u4ee5\u83b7\u5f97\u5408\u5e76\u671f\u95f4\u8981\u4f7f\u7528\u7684\u5806\u7a7a\u95f4\u7684\u6700\u7ec8\u6bd4\u4f8b<\/p>\n<h4 id=\\\"withMaxMemorySizePerCompactionInBytes\\\">withMaxMemorySizePerCompactionInBytes(maxMemorySizePerCompactionInBytes = 1GB)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.memory.compaction.fraction<br \/>\nHoodieCompactedLogScanner\u8bfb\u53d6\u65e5\u5fd7\u5757\uff0c\u5c06\u8bb0\u5f55\u8f6c\u6362\u4e3aHoodieRecords\uff0c\u7136\u540e\u5408\u5e76\u8fd9\u4e9b\u65e5\u5fd7\u5757\u548c\u8bb0\u5f55\u3002 \u5728\u4efb\u4f55\u65f6\u5019\uff0c\u65e5\u5fd7\u5757\u4e2d\u7684\u6761\u76ee\u6570\u53ef\u4ee5\u5c0f\u4e8e\u6216\u7b49\u4e8e\u76f8\u5e94\u7684parquet\u6587\u4ef6\u4e2d\u7684\u6761\u76ee\u6570\u3002\u8fd9\u53ef\u80fd\u5bfc\u81f4Scanner\u51fa\u73b0OOM\u3002 \u56e0\u6b64\uff0c\u53ef\u6ea2\u51fa\u7684\u6620\u5c04\u6709\u52a9\u4e8e\u51cf\u8f7b\u5185\u5b58\u538b\u529b\u3002\u4f7f\u7528\u6b64\u914d\u7f6e\u6765\u8bbe\u7f6e\u53ef\u6ea2\u51fa\u6620\u5c04\u7684\u6700\u5927\u5141\u8bb8inMemory\u5360\u7528\u7a7a\u95f4\u3002<\/p>\n<h4 id=\\\"withWriteStatusFailureFraction\\\">withWriteStatusFailureFraction(failureFraction = 0.1)<\/h4>\n<p>\u5c5e\u6027\uff1ahoodie.memory.writestatus.failure.fraction<br \/>\n\u6b64\u5c5e\u6027\u63a7\u5236\u62a5\u544a\u7ed9\u9a71\u52a8\u7a0b\u5e8f\u7684\u5931\u8d25\u8bb0\u5f55\u548c\u5f02\u5e38\u7684\u6bd4\u4f8b<br \/>\n<\/section>\n","protected":false},"excerpt":{"rendered":"<p>\u672c\u6587\u4e3b\u8981\u4ecb\u7ecd\u4e86\u51e0\u79cd\u914d\u7f6e\u5199\u5165\u6216\u8bfb\u53d6Hudi\u6570\u636e\u96c6\u7684\u4f5c\u4e1a\u7684\u65b9\u6cd5\u3002 \u7b80\u800c\u8a00\u4e4b\uff0c\u60a8\u53ef\u4ee5\u5728\u51e0\u4e2a\u7ea7\u522b\u4e0a\u63a7\u5236\u884c\u4e3a\u3002 Spar &hellip;<\/p>\n<p class=\"read-more\"> <a class=\"\" href=\"http:\/\/www.weijing.co\/?p=1442\"> <span class=\"screen-reader-text\">Hudi\u914d\u7f6e<\/span> \u67e5\u770b\u5168\u6587 &raquo;<\/a><\/p>\n","protected":false},"author":4,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"default","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"default","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":""},"categories":[116],"tags":[],"uagb_featured_image_src":{"full":false,"thumbnail":false,"medium":false,"medium_large":false,"large":false,"1536x1536":false,"2048x2048":false,"mailpoet_newsletter_max":false,"woocommerce_thumbnail":false,"woocommerce_single":false,"woocommerce_gallery_thumbnail":false,"shop_catalog":false,"shop_single":false,"shop_thumbnail":false},"uagb_author_info":{"display_name":"weijing","author_link":"http:\/\/www.weijing.co\/?author=4"},"uagb_comment_info":1,"uagb_excerpt":"\u672c\u6587\u4e3b\u8981\u4ecb\u7ecd\u4e86\u51e0\u79cd\u914d\u7f6e\u5199\u5165\u6216\u8bfb\u53d6Hudi\u6570\u636e\u96c6\u7684\u4f5c\u4e1a\u7684\u65b9\u6cd5\u3002 \u7b80\u800c\u8a00\u4e4b\uff0c\u60a8\u53ef\u4ee5\u5728\u51e0\u4e2a\u7ea7\u522b\u4e0a\u63a7\u5236\u884c\u4e3a\u3002 Spar&hellip;","_links":{"self":[{"href":"http:\/\/www.weijing.co\/index.php?rest_route=\/wp\/v2\/posts\/1442"}],"collection":[{"href":"http:\/\/www.weijing.co\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.weijing.co\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.weijing.co\/index.php?rest_route=\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"http:\/\/www.weijing.co\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1442"}],"version-history":[{"count":0,"href":"http:\/\/www.weijing.co\/index.php?rest_route=\/wp\/v2\/posts\/1442\/revisions"}],"wp:attachment":[{"href":"http:\/\/www.weijing.co\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1442"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.weijing.co\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1442"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.weijing.co\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1442"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}