> On Dec. 4, 2018, 10:08 p.m., Bharathkrishna Guruvayoor Murali wrote: > > Hi, the patch looks good, I just had a quick question. Does this mean that > > all the alters happen in one transaction? Will it prevent concurrent > > operations for the whole time a large number of partitions are altered?
No it will not. What was changed in this patch, that the list of partitions are passed to the metastore in one call, keeping the opened connections to minimum. Under the hood everything stayed the same, so it is possible to perform concurrent operations. - Laszlo ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/69341/#review211026 ----------------------------------------------------------- On Nov. 30, 2018, 11:31 a.m., Laszlo Pinter wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/69341/ > ----------------------------------------------------------- > > (Updated Nov. 30, 2018, 11:31 a.m.) > > > Review request for hive. > > > Repository: hive-git > > > Description > ------- > > HIVE-20891: Call alter_partition in batch when dynamically loading partitions > > > Diffs > ----- > > > ql/src/java/org/apache/hadoop/hive/metastore/SynchronizedMetaStoreClient.java > e8f362357537e73502f743a9df189dec9be2da5d > ql/src/java/org/apache/hadoop/hive/ql/metadata/Hive.java > e185bf49d42da9d1497643c20bbd71edaf071bf1 > > > Diff: https://reviews.apache.org/r/69341/diff/3/ > > > Testing > ------- > > > Thanks, > > Laszlo Pinter > >
