摘要:
According to the embodiments, a first storage area and a second storage area specified by a trim request is managed by a first management unit, and the second storage area specified by the trim request is managed by a second management unit. A block in which data of the first management unit are all specified by the trim request from the first or second storage areas and a block in which data of the second management unit are all specified by the trim request from the second storage area are released.
摘要:
A WC resource usage is compared with an auto flush (AU) threshold Caf that is smaller than an upper limit Clmt, and when the WC resource usage exceeds the AF threshold Caf, the organizing state of a NAND memory 10 is checked. When the organizing of the NAND memory 10 has proceeded sufficiently, data is flushed from a write cache (WC) 21 to the NAND memory 10 early, so that the response to the subsequent write command is improved.
摘要:
According to one embodiment, a memory system includes a data manager and a data restorer. The data manager multiplexes difference logs by a parallel writing operation and stores them in a second storage area, the difference logs being difference logs indicating difference information before and after update of a management table; and thereafter multiplexes predetermined data as finalizing logs and stores them in the second storage area. The data restorer determines a system status at startup of the memory system, by judging whether irregular power-off occurs or data destruction occurs in the second storage area, based on a data storage state of the difference logs and the finalizing logs stored in the second storage area.
摘要:
According to one embodiment, a free blocks included in a nonvolatile semiconductor memory are classified into a plurality of free block management lists. When a free block is acquired at normal priority, the free block is acquired from the free block management list in which a number of free blocks is larger than a first threshold. When a free block is acquired at high priority, the free block is acquired from the free block management list irrespective of the first threshold.
摘要:
Process migration method includes copying first process context indicative of first processing, transmitting process context to second computer, causing first computer to start generation of first execution record, causing second computer to receive process context, determining, from first execution record, whether first processing should be migrated, if it is determined that first processing should postpone being migrated, finishing generation of first execution record, starting generation of second execution record, transmitting first execution record to second computer, reproducing process context, and determining, from second execution record, whether first processing should be migrated, after reproducing of process context is finished in the second computer.
摘要:
A forward lookup address translation table and a reverse lookup address translation table stored in a nonvolatile second storing unit are transferred as a master table to a volatile first storing unit at a time of start-up. When an event occurs so that the master table needs to be updated, difference information before and after update of any one of the forward lookup address translation table and the reverse lookup address translation table is recorded in the first storing unit as a log, thereby reducing an amount of the log.
摘要:
A storage device for which restoration of data is possible without the need for excess equipment includes mirrored storage units for online use and for backup, and is capable of performing a data backup while continuing input and output of data. The storage device is further provided with an update data storage unit for storing update data. According to the storage device, even if the information stored in a storage unit for online use is not capable of being read because of a fault in the storage unit, it is possible to restore the data in the backup storage unit to the state that existed in the storage unit for online use immediately before the fault by applying the update data to the information stored in the storage device for backup.
摘要:
A controller sets, out of a data range that is specified in a read request from a host device, a predetermined size of a first data range that follows a top portion of the data range and a predetermined size of a second data range that follows the first data range, and after transfer, to the host device, of data corresponding to the first data range from a second storage unit or a third storage unit having smaller data output latency than the first storage unit in which read/write of data is performed is started, the controller searches for data corresponding to the second data range in the second storage unit or the third storage unit.
摘要:
An information processing system periodically performs a real-time operation including a plurality of chained tasks. The system includes a plurality of processors, a unit for dividing the chained tasks into a first task group and a second task group based on a relationship in order of execution among the tasks, the second task group being executed after the first task group, and a unit for performing a scheduling operation of periodically assigning each of the first task group and the second task group to at least one of the processors to periodically execute the first task group at regular time intervals and periodically execute the second task group at the regular time intervals with a one-period delay relative to the first task group.
摘要:
An information processing system performs a real-time operation including a combination of a plurality of tasks. The system includes a plurality of processors, a unit which stores structural description information and a plurality of programs describing procedures corresponding to the tasks, the structural description information indicating a relationship in input/output between the programs and including cost information concerning time required for executing each of the programs, a unit which determines an execution start timing and execution term of each of a plurality of threads for execution of the programs based on the structural description information, and a unit which performs a scheduling operation of assigning the threads to at least one of the processors according to a result of the determining.