Long term archiving and storage systems
DISC research projects
The high-performance computing (HPC) community has long needed a scalable,
global, and secure file system to store and access data efficiently.
Due to the extremely large volumes of data generated in HPC environments,
a hierarchical archival storage system is a necessity.
The data bandwidth that will be needed between levels of the storage
hierarchy is expected to exceed 350 TB/hr in 2006.
Existing systems, however, have demonstrated capabilities of only
10.1 TB/hr for file-level backup and 4.5 TB/hr for file-level restore.
Concurrent with the exploding bandwidth and storage requirements of the
HPC community, businesses of all sizes also are demanding scalable,
reliable, and secure storage to archive increasingly large amounts of data.
Retaining and protecting an enterprise's information assets is vital for
conducting business and often is required by law and government mandates,
such as the Sarbanes-Oxley Act, HIPAA, DOD 5015.2-STD,
and the European Data Privacy Directive.
To achieve the required level of security and privacy,
data must be encrypted both at rest and in transit.
However, the need to protect and preserve cryptographic keys
in a long-term archive presents a major challenge
since many unforeseen changes can occur during the data's lifetime.
This project is developing a high-performance long-term data management
system that will ensure the necessary levels of security throughout
the data's lifecycle.
This hierarchical cluster-based archival storage solution includes:
1) transparent backup, restore, and data access operations that will
allow individual application programs and business entities to securely
and efficiently archive data for decades;
2) high-performance data access in a cluster computing environment;
and 3) innovative techniques for efficiently ensuring long-term data security and accessibility, including long-term key management.
Return
to: U of M Home
