Hi, With the help of Yuantian, Zesong Sun and Lei Rui, the refactor of query process takes a step forward: vectorizing the raw data query. Indeed there is a lot of work to do, I suggest to merge this PR to master first because we change a lot. After merging, other optimization could be split into small PRs.
The PR is https://github.com/apache/incubator-iotdb/pull/652 , welcome to review. Best, Jialin Qiao Jialin Qiao (Jira) <[email protected]> 于2019年12月24日周二 上午9:24写道： > Jialin Qiao created IOTDB-386: > --------------------------------- > > Summary: Vectorize the raw data query process > Key: IOTDB-386 > URL: https://issues.apache.org/jira/browse/IOTDB-386 > Project: Apache IoTDB > Issue Type: Improvement > Reporter: Jialin Qiao > > > Currently, the readers in IoTDB most follows a tuple-at-a-time iterator > mode. It's better to replace it with column-at-a-time. > > The related academic research is "C-Store: A Column-oriented DBMS". > > !image-2019-12-24-09-21-19-827.png|width=326,height=194! > > > > > > -- > This message was sent by Atlassian Jira > (v8.3.4#803005) > -- ————————————————— Jialin Qiao School of Software, Tsinghua University 乔嘉林 清华大学 软件学院
