[ https://issues.apache.org/jira/browse/IOTDB-5273?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17655866#comment-17655866 ]
Wenwei Shu commented on IOTDB-5273: ----------------------------------- 测试结果:在master 656d281上可以复现问题,使用第一个配置文件运行benchmark关闭合并生成数据,再开启合并,最终有2个乱序TsFile残留,在更新版本master 0f944f7上测试已经无法复现问题 > [fast compaction]The performance is slow ,there are out-of-order tsfiles > after compaction > ----------------------------------------------------------------------------------------- > > Key: IOTDB-5273 > URL: https://issues.apache.org/jira/browse/IOTDB-5273 > Project: Apache IoTDB > Issue Type: Bug > Components: mpp-cluster > Affects Versions: master branch > Reporter: 刘珍 > Assignee: Wenwei Shu > Priority: Major > Attachments: 1_luanxu.conf, 2_luanxu.conf, 3_luanxu.conf, > 4_luanxu.conf, image-2022-12-23-18-10-25-061.png > > Original Estimate: 96h > Remaining Estimate: 96h > > master 1222_656d281 > 问题描述: > 私有云1期,周测乱序配置,fast合并性能慢,且合并完成依然有乱序tsfile。 > !image-2022-12-23-18-10-25-061.png|width=979,height=481! > 测试环境 > 1. 私有云1期 > 关合并,生成数据。 > 配置文件见附件 > 2.对比合并性能,无其他读写操作。 > ConfigNode配置 > MAX_HEAP_SIZE="2G" > DataNode配置 > MAX_HEAP_SIZE="18G" > MAX_DIRECT_MEMORY_SIZE="6G" > Common配置 > time_partition_interval=60480000000 > compaction_io_rate_per_sec=1000 -- This message was sent by Atlassian Jira (v8.20.10#820010)