Long-term Storage at RZG

The Computing Center Garching (RZG) offers to all Max Planck Institutes a long-term storage service. Through redundancy at the logical and at the physical level, the storage of important data like historical documents, medical records and other scientific data can be guaranteed for a virtually unlimited period of time.

 

What is Long-term Storage?

Long-term Storage is the storage of data for a period of time which surpasses the typical life-cycle of the hardware and software components employed, with the goal of keeping the data forever.

This typical life-cycle is:

• for archive hardware (tape and disk systems): about 5 years
• for HSM systems (software and hardware): about 10 years
• for filesystems with HSM functionality: about 20 years
• for metadata servers (databases): about 20 years
• for document formats (.tif, .pdf ...): ??? years

Replacing the components used by copying the data can take years. That is why short-lived components must be hidden whenever possible:

• Hardware under HSM systems
• HSM Systems under filesystems with HSM functionality or under a database driven portal

 

Challenges and expertise

Keeping pace with the storage technology changes, migrating old data to new systems and managing long-term storage data as a whole is a task which easily overcharges smaller computing centers. The Computing Center Garching can look back at over 35 years of experience with different generations of archive systems. Throughout this years, the RZG has developed a solid expertise in long-term storage which it can offer to all MPIs.

   History of HSM systems at RZG

 

Real-world scenarios

Data from different Max Planck Institutes has been stored at RZG since the early 1980's.

   Some examples of the long-term archive at RZG

 

Best practices

A description of the archiving systems available at RZG to help the user decide which one to use.