摘要:
Provided is a hot data management method based on hit counter, which prevents loads from being concentrated on a specific data server due to hot data by efficiently managing the hot data in the asymmetric storage system. Each of data servers of an asymmetric storage system checks hit counter for data stored in the corresponding data server to maintain hit counter information on the data in a latest state. The data server transmits the hit counter information to a metadata server at predetermined intervals. The data server makes or deletes a copy of the data according to the determination of the metadata server whether the data is hot data.
摘要:
Disclosed is a method of calculating parity. The method dividing, by a client system, a file which is to be stored into chunks of a preset size and distributively storing the chunks in a plurality of data servers; sending, by the client system, a data write complete notification message to a metadata server; storing, by the metadata server, a file sent through the data write complete notification message in a recovery queue; and sending, by the metadata server, a random data server a parity calculation request for the file stored in the recovery queue.
摘要:
Provided is a hot data management method based on hit counter, which prevents loads from being concentrated on a specific data server due to hot data by efficiently managing the hot data in the asymmetric storage system. Each of data servers of an asymmetric storage system checks hit counter for data stored in the corresponding data server to maintain hit counter information on the data in a latest state. The data server transmits the hit counter information to a metadata server at predetermined intervals. The data server makes or deletes a copy of the data according to the determination of the metadata server whether the data is hot data.
摘要:
The present invention relates to an apparatus and method of performing an error recovery process in an asymmetric clustering file system that has higher efficiency in data recovery when a data server in an asymmetric clustering file system fails than a method of processing the data recovery in the metadata server. The present invention includes receiving a chunk list requiring recovery by a data server included in the other data server groups than a data server group including a failed data server among a plurality of data server groups, requesting chunk data necessary for recovering an erroneous chunk from a data server in the other data server groups to the other data servers than the failed data server in the data server group, and recovering the erroneous chunk based on the chunk data by the data server in the other data server group.
摘要:
The present invention relates to an apparatus and method of performing an error recovery process in an asymmetric clustering file system that has higher efficiency in data recovery when a data server in an asymmetric clustering file system fails than a method of processing the data recovery in the metadata server. The present invention includes receiving a chunk list requiring recovery by a data server included in the other data server groups than a data server group including a failed data server among a plurality of data server groups, requesting chunk data necessary for recovering an erroneous chunk from a data server in the other data server groups to the other data servers than the failed data server in the data server group, and recovering the erroneous chunk based on the chunk data by the data server in the other data server group.
摘要:
Disclosed are an apparatus, a system, and a method for recovering meta data using fragmentary information. The system for recovering meta data includes: a data server storing recovery information for recovering at least one data file and meta data related to each of the data file; and a meta data server storing the meta data related to the data files and receiving the recovery information from the data server when the meta data are lost to recover the lost meta data. By this configuration, the exemplary embodiment of the present invention stably can the lost meta data of the meta data server by using the recovery the meta data and prevent the phenomenon that the meta data collides with each other on the same directory hierarchical structure.
摘要:
Provided are an apparatus and a method which can be easily implemented with flexibility enabling distributing all metadata of trees and files in an asymmetric distributed file system. The apparatus includes: a metadata storage unit storing metadata corresponding to a part of partitions of a virtual metadata address space storing metadata for directories and/or files for each of the partitions; and a metadata storage management unit controlling the metadata so that the metadata are stored in the metadata storage unit and manages a master map including information on the part of the partitions. Since all directories and files can be distributed to a plurality of metadata servers without a limitation, it is possible to prevent a load from being concentrated on a predetermined metadata server. Metadata roles of the metadata servers are very simply readjusted and as a result, the load can be easily distributed in a partition level.
摘要:
Disclosed is a cloud storage managing a plurality of files, including: a plurality of metadata servers managing a plurality of metadata associated with the plurality of files; a plurality of data servers managing the data of the plurality of files; and a cluster management server managing the plurality of metadata servers and the plurality of data servers.
摘要:
A method and system for processing sensor data in a context-aware system is provided. The system includes: a sensor service collecting the sensor data from a driver of a physical sensor and generating a low level event; an STHQ temporarily storing the generated event and an event generated previously by an event interceptor; and a sensor framework comprised of the event interceptor forming an EIC (event interpretation chain) which interprets the generated event and the event stored in the STHQ and generated previously in connection with each other, and generating or deleting an event or processing the generated event as an event necessary for an upper level application service through an interlocking with the STHQ. The present invention provides active services on the basis of context knowledge inputted from various sensors.
摘要:
A remote power management technology for a cluster system of the present invention can monitor and control the power status of multiple hosts of the cluster system individually, thereby decreasing power consumption of the cluster system. A remote power management system includes a plurality of management nodes, a power proxy server, and a power management server. The management nodes, which manage power of the hosts of the cluster system respectively, are divided into groups. The power proxy server manages each group, monitors the power status of each management node of the group to generate proxy monitoring information, and transmits a power setting command to the management node of the specific host requiring power setting. The power management server monitors the power status of the hosts using the proxy monitoring information and transmits the power setting command to the power proxy server of the group including the management node of the specific host.