Peng Yu, on Tuesday, October 15, 2019 06:47 PM, wrote... > > Hi, > > I'd like to use sqlite3 db files on many compute nodes. But they > should access the same storage device for the sqlite3 db files. The > directory storing the db files looks the same on any compute node > logically---the storage is mounted at the same mount point on the > compute nodes. > > To achieve this, I think a network file system (could be distributed > (parallel) file system). But I know NFS is very bad at supporting > sqlite3. Many other distributed (parallel) file systems also are not > designed to support sqlite3, either. > > Is there a solution that are known to fill in this niche? Thanks.
I hope you have read the Appropriate Uses for SQLite[1], and have understood that whole page. If you have, then if your network is good, and you are not going to have multiple writers, then you should be good with any system as long as your network is properly set and kept. We have an 8 writers, yes with some specialized code to keep track of these writers, and maintain a healthy database. This program has been working since 2007, at the beginning with some major problems (database corruption), but since 2010, after some major code implementation to manage the writers, it has been working great. Take this with a grain of salt, or with some salted peanuts, but just know that it can work properly if the writers are handled properly. Just thinking out-loud. josé [1] https://www.sqlite.org/whentouse.html _______________________________________________ sqlite-users mailing list sqlite-users@mailinglists.sqlite.org http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users