Abstract: Embodiments of the present disclosure relate to a method, a device and a computer program product for managing a distributed system. The method comprises sending heartbeat messages from a master node to a plurality of slave nodes, the master node and the plurality of slave nodes being included in a plurality of nodes in the distributed system, and the plurality of nodes being divided into one or more partitions. The method further comprises, in response to receiving a response to the heartbeat messages from a portion of slave nodes in the plurality of slave nodes, determining respective states of the one or more partitions. In addition, the method further comprises a state of a first slave node in the plurality of slave nodes at least based on the respective states of the one or more partitions, the master node failing to receive a response to the heartbeat messages from the first slave node.

Abstract: Embodiments of the present disclosure provide a method, device and computer program product for flushing metadata in a multi-core system. The method comprises: moving a metadata identifier included in a sub-list of a first list to a corresponding sub-list of a second list, the sub-list of the first list and the corresponding sub-list of the second list being associated with the same processing unit; moving the metadata identifier from the corresponding sub-list of the second list to a third list based on a storage position of the metadata identifier; and determining metadata to be flushed from the third list to a metadata repository. By means of the method and device for flushing metadata as proposed in the present disclosure, metadata synchronization contention can be reduced, IO efficiency can be improved, response time can be decreased, and the cache hit can be increased.

Abstract: Embodiments of the present disclosure provide a method, device and computer program product for flushing metadata in a multi-core system. The method comprises: moving a metadata identifier included in a sub-list of a first list to a corresponding sub-list of a second list, the sub-list of the first list and the corresponding sub-list of the second list being associated with the same processing unit; moving the metadata identifier from the corresponding sub-list of the second list to a third list based on a storage position of the metadata identifier; and determining metadata to be flushed from the third list to a metadata repository. By means of the method and device for flushing metadata as proposed in the present disclosure, metadata synchronization contention can be reduced, IO efficiency can be improved, response time can be decreased, and the cache hit can be increased.

Abstract: Embodiments of the present disclosure relate to a method, a device and a computer program product for managing a distributed system. The method comprises sending heartbeat messages from a master node to a plurality of slave nodes, the master node and the plurality of slave nodes being included in a plurality of nodes in the distributed system, and the plurality of nodes being divided into one or more partitions. The method further comprises, in response to receiving a response to the heartbeat messages from a portion of slave nodes in the plurality of slave nodes, determining respective states of the one or more partitions. In addition, the method further comprises a state of a first slave node in the plurality of slave nodes at least based on the respective states of the one or more partitions, the master node failing to receive a response to the heartbeat messages from the first slave node.