Re: [Discuss] IntervalJoin one side sorted cache

2020-03-06 Thread Chen Qin
r could tolerant clear up state earlier, > why not just change [lowerbound, upbound] shorter directly? > Looking forward to your response, thanks. > > > ---------- > 发件人:Chen Qin > 发送时间:2020年3月6日(星期五) 10:54 > 收件人:dev > 主

回复:[Discuss] IntervalJoin one side sorted cache

2020-03-06 Thread 张静(槿瑜)
nfused. If user could tolerant clear up state earlier, why not just change [lowerbound, upbound] shorter directly? Looking forward to your response, thanks. -- 发件人:Chen Qin 发送时间:2020年3月6日(星期五) 10:54 收件人:dev 主 题:[Discuss] IntervalJoi

[Discuss] IntervalJoin one side sorted cache

2020-03-05 Thread Chen Qin
Hi there, I would like kick off discussion on https://issues.apache.org/jira/browse/FLINK-16392 and discuss what is best way moving forward. Here is problem statement and proposal we have in mind. Please kindly provide feedback. Native intervaljoin rely on statebackend(e.g rocksdb) to insert/fetc