[ https://issues.apache.org/jira/browse/HDFS-17388?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
ZanderXu updated HDFS-17388: ---------------------------- Summary: [FGL] Client RPCs involving write process supports fine-grained lock (was: [FGL] Create RPC supports this fine-grained locking I) > [FGL] Client RPCs involving write process supports fine-grained lock > -------------------------------------------------------------------- > > Key: HDFS-17388 > URL: https://issues.apache.org/jira/browse/HDFS-17388 > Project: Hadoop HDFS > Issue Type: Sub-task > Reporter: ZanderXu > Assignee: ZanderXu > Priority: Major > Labels: pull-request-available > > The create RPC just involves directory tree if it creates a new file, and > most scenes are like this. It involves blocks only if the file is existing > and it tries to overwrite it. > So in most scenarios, the create RPC just needs FS lock. > The current implementation just holds the global write lock, so in order for > the improvement to be better accepted, the first step is just to replace the > lock mode without changing logic. We can minimize the scope of the global BM > lock in the second step. -- This message was sent by Atlassian Jira (v8.20.10#820010) --------------------------------------------------------------------- To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org