The high-performance data storage systems at NCI consist of multiple powerful filesystems with some of the highest performance in the country.
Tightly integrated with a collocated supercomputer across a dedicated high-speed network, NCI’s more than 100 petabytes of data storage brings high-performance data to the fore and makes groundbreaking big data research possible.
NCI’s simple, efficient and effective data storage systems provide rapid access to petabytes of research-ready data from across our computing systems.
Active project data at NCI is stored on persistent Lustre filesystems providing world-class performance and data security. This makes data processing and analysis much easier and more efficient, as the computers have high-speed access to entire data collections during their operation. This also helps improve collaboration and data sharing, as researchers can access and view datasets in-situ at NCI, rather than by transferring files around the country. NCI’s data storage systems simplify the research task by allowing scientists to focus on their work and not on managing their data.
In total, NCI stores over 100 petabytes of research data. Around 50 petabytes of active project data is stored in the five global filesystems, and a further 50 petabytes of archival data is stored on highly durable magnetic tape. The archival data includes collections of satellite imagery and environmental photography going back to the 1980s.
NCI’s data storage is growing at a rate of over 15 terabytes per day. As the computational and data needs of the research community grow, so does the need to maintain and support the necessary storage systems into the future.
For detailed specifications of NCI’s data storage systems, please see the dedicated page here.