Disk areas

Similar to your own computer, also supercomputers have different directories readily available. Their use and names are a bit different though:

Puhti disk areas

Name	Access	Path	Cleaning	Capacity	Number of files	Use
home	Personal	/users/cscusername	No	10 GiB	100 000 files	personal settings and files
projappl	Project	/projappl/project_200xxxx	No	50 GiB	100 000 files	installation files
scratch	Project	/scratch/project_200xxxx	180 days	1 TiB	1 000 000 files	main working area

Temporary fast disks

• CSC Docs: Login node local tmp $TMPDIR for compiling, cleaned frequently.

• CSC Docs: NVMe - $LOCAL_SCRATCH in batch jobs,

  • NVMe is accessible only during your job allocation, inc. interactive job

  • You must copy data in and out during your batch job

  • If your job reads or writes a lot of small files, using this can give 10x performance boost

Avoid unneccesary reading and writing

Avoid unnecessary reads and writes of data to improve I/O performance

• Read and write in big chunks and avoid reading/writing lots of small files

  • If unavoidable, use fast local NVMe disk, not Lustre (i.e. /scratch)

LUMI disk areas

• LUMI docs: storage

• LUMI has similar main disks, but different temporary disks.

Disk status

• Display usage and quota of all your disk areas: csc-workspaces

• Display the amount of data and number of files within a given folder: LUE

• CSC Docs: LUE tutorial

module load lue
lue --display-level=2 /scratch/project_200xxxx/

path, total size, in dir size, % of total, % of dir
---------------------------------------------------
/scratch/project_200xxxx/dirA        8.4GB  356KB 100.0 100.0
    results                            3.7GB  458MB 44.15 44.15
        simu1                              2.8GB  522MB 32.84 74.38 NOSIZE:1
        simu2                              521MB  521MB 6.02  13.64 NOSIZE:1
    installation                       1.4GB  48KB  16.2  16.2 
        gcc10                              351MB  351MB 4.05  25.02
        clang15                            351MB  351MB 4.05  25.02
        intel                              350MB  350MB 4.04  24.94

Some best practice tips

• Take backups of important files. Data on Puhti disks is not backed up.

  • Allas is best CSC option for back-up.

  • Github or similar for code.

• Supercomputer disks do not work well with too many small files (see the file limits above)

  • Plan your analysis in a way that too many files are not needed.

  • Keep the small files in one zip-file, unzip it only on local fast disks during the analysis.

  • Don’t create a lot of files in one folder

• CSC Docs: Best practice performance tips for using Lustre

• Databases:

  • Only file databases (SQLite, GeoPackage) can be kept in supercomputer disks.

  • For PostgreSQL (but not PostGIS) use CSC CSC Docs: Database-as-service.

  • For any other database set up virtual machine in cPouta.