Hi, here are my opinions:
1. the Leader of each DataGroup should sync data to remote, and ask the destination node of the data file before sending the data. 2. the remote IoTDB tells the destination of the data file to the edge IoTDB. 3. the remote Leader of the DataGroup receive the data file and sync to the follower. And, I think we could redesign the sync tool to support sync data to the formats of other storage system. This will enrich our ecosystem. Thanks. Chao Wang BONC Ltd From: Xiangdong Huang Date: 2020-09-29 11:50 To: dev CC: notifications Subject: Re: [jira] [Created] (IOTDB-923) [Distributed] Sync tool to support edge send data to cloud Hi, Any ideas? Send to some node and let the node partition the data file and then forward to other nodes? ----------------------------------- Xiangdong Huang School of Software, Tsinghua University 黄向东 清华大学 软件学院 WangChao (Jira) <[email protected]> 于2020年9月28日周一 上午10:15写道： > WangChao created IOTDB-923: > ------------------------------ > > Summary: [Distributed] Sync tool to support edge send data to > cloud > Key: IOTDB-923 > URL: https://issues.apache.org/jira/browse/IOTDB-923 > Project: Apache IoTDB > Issue Type: New Feature > Components: Core/Cluster > Reporter: WangChao > > > Now we have sync tool to support edge send data to cloud in single mode. > > We also need support sync tool in cluster version. > > > > -- > This message was sent by Atlassian Jira > (v8.3.4#803005) >
