The Center for Research in Storage Systems (CRSS) conducts research in next-generation data storage systems, paving the way for companies to build and maintain systems that provide secure, reliable, energy-efficient, and cost-effective data storage. CRSS research focuses on the integration of new technologies into the storage stack and on the use of new computational methods and system-building techniques to improve storage system performance, usability, reliability, and security. The mission of CRSS is to improve the manageability, scalability, security, reliability, longevity, and performance of data storage systems. CRSS is a partnership between academia and industry to explore and develop new technologies and methods. CRSS also facilitates collaboration in research and education, providing pathways to simplify direct transfer of university-developed ideas, research results, and technology to its industrial sponsors. This enables the industrial members to be more competitive in the global marketplace. CRSS also focuses on providing talented graduate and undergraduate students the technical background and industry interaction required to become the next generation of scientists and engineers.
Research Areas
Application of machine learning to improve storage system performance
Explore machine-learning techniques for system trace generation and analysis.
Investigate techniques to use machine learning to optimize system design in a running system.
Data storage security
Develop techniques to secure data against snooping by hiding it in less-secure data.
Develop ways to quickly and securely delete data to satisfy requirements from initiatives such as Europe’s General Data Protection Regulation and California's data privacy laws. These techniques may lead to better approaches for data storage that are secure against snooping, even by untrusted cloud storage providers.
Efficient utilization of byte-addressable nonvolatile memory (NVM), such as Intel Optane, into the storage stack
Explore multiple techniques for supporting NVM persistence across crashes.
Investigate changes to the operating system to better integrate NVM with lower overhead and higher security.
Long-term data storage
Develop techniques to make long-term data storage more reliable and secure.
Investigate the impacts of different storage technologies — flash, shingled disk, DNA storage, glass, and others — on the long-term viability and cost of archival data storage.
Investigate techniques to better leverage long-term data storage by making it easier to run analyses on “cold” data.
Scalable storage systems
Develop approaches that leverage high-bandwidth data center interconnects to provide scalable, reliable flash storage at ultrahigh speeds for use in cloud data center computing.
Facilities & Resources
Partner Organizations
Abbreviation |
CRSS
|
Country |
United States
|
Region |
Americas
|
Primary Language |
English
|
Evidence of Intl Collaboration? |
|
Industry engagement required? |
Associated Funding Agencies |
Contact Name |
Ethan Miller
|
Contact Title |
Center Director
|
Contact E-Mail |
elm@cs.ucsc.edu
|
Website |
|
General E-mail |
|
Phone |
|
Address |
The Center for Research in Storage Systems (CRSS) conducts research in next-generation data storage systems, paving the way for companies to build and maintain systems that provide secure, reliable, energy-efficient, and cost-effective data storage. CRSS research focuses on the integration of new technologies into the storage stack and on the use of new computational methods and system-building techniques to improve storage system performance, usability, reliability, and security. The mission of CRSS is to improve the manageability, scalability, security, reliability, longevity, and performance of data storage systems. CRSS is a partnership between academia and industry to explore and develop new technologies and methods. CRSS also facilitates collaboration in research and education, providing pathways to simplify direct transfer of university-developed ideas, research results, and technology to its industrial sponsors. This enables the industrial members to be more competitive in the global marketplace. CRSS also focuses on providing talented graduate and undergraduate students the technical background and industry interaction required to become the next generation of scientists and engineers.
Abbreviation |
CRSS
|
Country |
United States
|
Region |
Americas
|
Primary Language |
English
|
Evidence of Intl Collaboration? |
|
Industry engagement required? |
Associated Funding Agencies |
Contact Name |
Ethan Miller
|
Contact Title |
Center Director
|
Contact E-Mail |
elm@cs.ucsc.edu
|
Website |
|
General E-mail |
|
Phone |
|
Address |
Research Areas
Application of machine learning to improve storage system performance
Explore machine-learning techniques for system trace generation and analysis.
Investigate techniques to use machine learning to optimize system design in a running system.
Data storage security
Develop techniques to secure data against snooping by hiding it in less-secure data.
Develop ways to quickly and securely delete data to satisfy requirements from initiatives such as Europe’s General Data Protection Regulation and California's data privacy laws. These techniques may lead to better approaches for data storage that are secure against snooping, even by untrusted cloud storage providers.
Efficient utilization of byte-addressable nonvolatile memory (NVM), such as Intel Optane, into the storage stack
Explore multiple techniques for supporting NVM persistence across crashes.
Investigate changes to the operating system to better integrate NVM with lower overhead and higher security.
Long-term data storage
Develop techniques to make long-term data storage more reliable and secure.
Investigate the impacts of different storage technologies — flash, shingled disk, DNA storage, glass, and others — on the long-term viability and cost of archival data storage.
Investigate techniques to better leverage long-term data storage by making it easier to run analyses on “cold” data.
Scalable storage systems
Develop approaches that leverage high-bandwidth data center interconnects to provide scalable, reliable flash storage at ultrahigh speeds for use in cloud data center computing.