Abstract:
A method and system for achieving highly available, fault-tolerant execution of components in a distributed computing system, without requiring the writer of these components to explicitly write code (such as entity beans or database transactions) to make component state persistent. It is achieved by converting the intrinsically non-deterministic behavior of the distributed system to a deterministic behavior, thus enabling state recovery to be achieved by advantageously efficient checkpoint-replay techniques. The method comprises: adapting the execution environment for enabling message communication amongst and between the components; automatically associating a deterministic timestamp in conjunction with a message to be communicated from a sender component to a receiver component during program execution, the timestamp representative of estimated time of arrival of the message at a receiver component. At a component, tracking state of that component during program execution, and periodically checkpointing the state in a local storage device. Upon failure of a component, the component state is restored by recovering a recent stored checkpoint and re-executing the events occurring since the last checkpoint. The system is deterministic by repeating the execution of the receiving component by processing the messages in the same order as their associated timestamp.
Abstract:
PROBLEM TO BE SOLVED: To provide a storage device which reduces the load of processing for detecting a mistake in writing while maintaining high reliability.SOLUTION: A generation code storage part 122 stores a generation code indicative of an update state of block data associatively with identification information on the block data. A generation code management part 13 stores generation codes in the generation code storage part 122 while allocating a new generation code to detected successive data and new generation codes to respective block data other than the successive data included in write data. A data write part 15 adds the new generation codes to the block data and successive data and writes them to a storage device 19. A determination part 17 determines whether the generation code added to read-out block data or successive data matches the read-out generation code stored in the generation code storage part 122. A data transmission part 18 transmits the read-out block data or successive data when the generation code match each other.
Abstract:
PROBLEM TO BE SOLVED: To provide a data processing technology for a disk array device capable of realizing data duplication for improvement in device performance while reducing writing processing to a cache memory (via a switch). SOLUTION: In this disk array device, a host interface part 11 is provided with a non-volatile memory part 22 storing writing data from a host computer/server 1 and a data transfer control part 21 transferring writing data from the host computer/server 1 to the non-volatile memory part 22 and a global cache memory part 14. In receipt of a writing request from the host computer/server 1, the writing data are transferred by the data transfer control part 21 from the host computer/server 1 to the non-volatile memory part 22 and the global chace memory part 14 via a switch part 15 to be written. COPYRIGHT: (C)2006,JPO&NCIPI
Abstract:
Restoring a mass storage device, including the corresponding data blocks stored thereon, to a state in which it existed at a prior instant in time to minimize the data loss caused by data blocks becoming corrupt or lost. After a mirrored or backup copy has been made, data blocks that are to be overwritten in response to a write request are stored in a preservation memory prior to being overwritten. The data blocks stored in the preservation memory are time-stamped to designate the chronological order by which the data blocks were overwritten. If data becomes corrupted, the data blocks of the preservation memory are applied to the corrupted data in reverse chronological order until such time that a valid, non-corrupted set of data is obtained. In this manner, data more recent than that associated with the full mirrored or backup copy can be reconstructed.
Abstract:
PROBLEM TO BE SOLVED: To provide a method and a device for constructing and executing a complicated software. SOLUTION: In a step S1200, input data are provided to one or plural defined storage positions. Next in a step S1300, it is determined which action of a dependent action system is to be made 'available'. When there is any action made available, control is moved to a step S1400. In the other case, control is jumped to a step S1500. In the step S1400, any one of available actions is executed.