Hi, No, I do not have a clear idea about that. Some minor improvements are: - get an optimal order of when reading multiple timeseries, e.g., for sql "select s1, s2, s3 ...", if the serialized data order is s1, s3 and s2, then change the physical plan to read s1, s3, s2 rather than s2, s3.. - rethinking all caches for querying. Allowing user to define the size of caches. etc..
Best, ----------------------------------- Xiangdong Huang School of Software, Tsinghua University 黄向东 清华大学 软件学院 Julian Feinauer <[email protected]> 于2019年8月19日周一 下午2:15写道： > Hi Xiangdong, > > thanks for bringing that up. > Your suggestion contains multiple interesting topics. > > As I'm active in Calcite I love Query Optimization. > What are the communities thoughts about that? > What do you plan to do or to enable? > > Julian > > Am 19.08.19, 05:10 schrieb "xiangdong Huang (JIRA)" <[email protected]>: > > xiangdong Huang created IOTDB-167: > ------------------------------------- > > Summary: v0.9.0 roadmap > Key: IOTDB-167 > URL: https://issues.apache.org/jira/browse/IOTDB-167 > Project: Apache IoTDB > Issue Type: Task > Reporter: xiangdong Huang > > > The v0.9.0 can have the following function: > # the merge of sequence files and unsequence files; > # the sync module > # the hadoop module > # query optimization > # define some subexceptions when executing SQL failed. > > Please add your suggestion in this thread. > > > > -- > This message was sent by Atlassian JIRA > (v7.6.14#76016) > > >
