PAVO: Petascale Storage for High Resolution Wide Area Video
Department of Defense
Defense Advanced Research Projects Agency
Agency Tracking Number:
Solicitation Topic Code:
Small Business Information
RNET TECHNOLOGIES, INC.
555 Bryant Street, #578, Palo Alto, CA, 94301
Socially and Economically Disadvantaged:
AbstractGigapixel cameras can generate petabits of data every hour, requiring vast bandwidth and storage in order to store the entire video sequence for offline forensic analysis. Realtime compression techniques will be developed to greatly reduce the storage requirements. Key requirements of the compression algorithm are computational power, such that the system is power and space efficient, high compression ratios and minimal artifacts. The compression algorithm will operate in a data parallel manner, allowing for very good scalability. RAID partitions will be used inorder to create large storage volumes, and to allow a configurable level of data reliability. Embedded processing elements (e.g., FPGA, FPOAs, GPUs) will be used to implement the compression algorithm in a cost and power efficient system.
* information listed above is at the time of submission.