Using nearline storage

This article is a draft

This is not a complete article: This is a draft, a work in progress that is intended to be published into an article, which may or may not be ready for inclusion in the main wiki. It should not necessarily be considered factual or authoritative.

Nearline is a filesystem virtualized onto tape

Nearline storage is like Project, except that files can be "virtualized" by moving them to tape. The files are still directly accessible, so this is a way to manage less-used files. On tape, they do not consume your disk quota, but can still be accessed if necessary.

This is useful because the capacity of our tape libraries is both large and expandable. When a file has been moved to tape (that is, "virtualized"), it will still appear in the nearline directory. If the virtual file is read, the reading process will block for some period, probably a few minutes, while the file contents are read from tape to disk. Then IO to the file will behave like a normal disk-based file.

Expected use

Because of the delay in reading from tape, Nearline is not intended to be used by jobs, where the delay would waste allocated time - it is only available on login and DTN nodes.

How to use

To use nearline, just put files into your ~/nearline/PROJECT directory.; after a period (currently 24 hours), they'll be stored on tape. If the file remains unchanged for another period (also 24h), the disk-based copy will be dropped (making the file virtualized on tape). Like most HPC storage, it's bad practice to have lots of small files. (In fact, presently small files won't be moved to tape at all.) So if you have large collections of small files, it's wise to bundle them using a tool like tar. Also note that if you remove a file in Nearline, the tape copy will be retained for up to 60 days (restoring such a file is like asking for recovery from backup - you have to email us the request.)