Data rebuild on feedback from a queue in a non-volatile solid-state storage

作者: Hari Kannan , Zhangxi Tan , Nenad Miladinovic , John Hayes , John D. Davis

DOI:

关键词:

摘要: A storage cluster is provided. The includes a plurality of nodes, each the nodes having nonvolatile solid-state memory and operations queues coupled to memory. configured distribute user data metadata throughout such that can access with failure two nodes. Each determine whether read 1 or more bits in via first path within latency budget. perform second path, responsive determination bit not

参考文章(238)
Wayne H. Vinson, Mark Allen Gaertner, Kevin Gomez, Ryan J. Goss, Jonathan W. Haines, Timothy R. Feldman, Data segregation in a storage device ,(2010)
Michael J. Uttormark, Michael J. Klemm, System and method for RAID management, reallocation and restriping ,(2007)
Narayanan Balakrishnan, Ajit Narayanan, Raja Jayaraman, Paresh Chatterjee, Maintaining data consistency in mirrored cluster storage systems using bitmap write-intent logging ,(2011)
Narayanan Balakrishnan, Raja Jayaraman, Paresh Chatterjee, Maintaining data consistency in mirrored cluster storage systems with write-back cache ,(2008)
Henry N. Jerez, Tolga Acar, Thomas Michael Roeder, Lan Duy Nguyen, Role-based distributed key management ,(2015)
Michael Jacobson, George Shin, Richard Brame, Communication in multiprocessor using proxy sockets ,(2011)
James Vash, Ugonna Echeruo, George Chrysos, Chyi-Chang Miao, Method and apparatus for implementing memory order models with order vectors ,(2004)