Thanks DaWei! -- Jialin Qiao School of Software, Tsinghua University
乔嘉林 清华大学 软件学院 > -----原始邮件----- > 发件人: "孙泽嵩" <[email protected]> > 发送时间: 2020-06-20 07:07:44 (星期六) > 收件人: [email protected] > 抄送: > 主题: Re: [Weekly Report] IoTDB Weekly News (2020-06-12~2020-06-19) > > Hi Dawei, > > Welcome back! > > Thanks for ur sharing : ) > > > Best, > ----------------------------------- > Zesong Sun > School of Software, Tsinghua University > > 孙泽嵩 > 清华大学 软件学院 > > > 2020年6月20日 01:23，Dawei Liu <[email protected]> 写道： > > > > 大家好， > > 很开心又一次与大家分享 IoTDB 社区周报。 > > > > > > [本周大事件] > > * 我们计划每2-3周举行一次在线讨论，以便让更多的贡献者能够深入交流。 > > * IoTDB 社区的成熟度评估(draft)已经发布，欢迎各位审阅[1]. > > * 0.10.0-RC4 正在投票，欢迎检查并发表意见和建议! > > * 新增加了 "show latest timeseries"功能, > > 该功能按照timeseries中最后一个点的时间戳降序输出，可以更方便地知道哪些时序还在写数据. > > * 优化了 MTree > > * 分享一个内存开销的实验结果，在MTree中保存2千万的时序元数据大约需要 7G 内存. > > > > > > [1]https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=148645763 > > > > > > > > [in-progress] > > * 新增 MTree 的检查点，加快当数据库中存在非常多的时序数据时的重启恢复速度。目前2千万数据大概需要花费2分钟。 > > * 新增对存储引擎、查询引擎、memTable 以及每个TsFile的索引内存控制。 > > * 新增基于时间范围删除数据的语句 > > > > > > > > > > [bug fix] > > [IOTDB-765] 修复当使用HDFS存储时，获取 upgrade.txt 文件产生的错误 > > [IOTDB-769] [To rel/0.10] 修复当使用PLAIN编码时，FLOAT/DOUBLE丢失精度的问题 > > [IOTDB-762] 修复设置versionInfo时产生的 IndexOutofBound 错误 > > [IOTDB-772] 优化 WinOS 下 start-server.sh 脚本中的 java 命令过长 > > 修复重启时死锁的bug > > [IOTDB-761]cherry-picked from v0.10: 修复 flink-tsfile-connecto 和 > > flink-example工程中的依赖冲突: "Some Enforcer rules have failed. Look above for > > specific messages explaining why the rule failed" > > Fix spark tsfile bug in 0.9.4/0.10.1 and 0.11.0 > > > > > > > > > > 祝各位周末愉快 > > > > > > Thanks > > ----- > > > > > > Dawei Liu > > > > > > > > > > On 06/20/2020 01:15，Dawei Liu<[email protected]> wrote： > > Dear community, > > > > > > > > I'd like to share the weekly report for you this week. > > > > > > [Big Event] > > * We plan to hold online-discussion every 2-3 weeks to let contributors > > communicate more deeply. > > * Maturity Evaluation draft of IoTDB is published, please have a review [1] > > * We are voting the 0.10.0-RC4 now, please check and vote! > > * A new feature: "show latest timeseries", for showing the latest active > > timeseries, i.e., ordering all timeseries according to how long the series > > has not written new data. > > * The MTree is optimized > > * Memory cost experiment, 20M time series metadata in MTree takes about 7G > > memory. > > > > > > [1] > > https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=148645763 > > > > > > [in-progress] > > * Check point of MTree to accelerate the restart process when there are > > many timeseries. It takes about 2 minutes for 20M timeseries. > > * Memory control of storage engine, query engine, flushing memtable, index > > of each tsfile > > * Support time range Deletion > > > > > > [bug fix] > > [IOTDB-765] Fix failed to get upgrade.txt file when using HDFS storage > > [IOTDB-769] [To rel/0.10] fix precision lost when using PLAIN for > > FLOAT/DOUBLE > > [IOTDB-762] Fix set version IndexOutofBound > > [IOTDB-772] The start-server.sh uses tool long java command in WinOS > > Fix Restart deadlock bug > > [IOTDB-761]cherry-picked from v0.10: flink-tsfile-connecto and > > flink-example dependency conflict: "Some Enforcer rules have failed. Look > > above for specific messages explaining why the rule failed" > > Fix spark tsfile bug in 0.9.4/0.10.1 and 0.11.0 > > > > > > > > > > Have a great weekend. > > > > > > Thanks > > —— > > > > > > Dawei Liu > > >
