摘要:
In response to the issue of a request for processing long data from an application node to a server A, the server A sends a message requesting a server B for the processing of the long data. The server B requests the application node to transmit the long data to the server B by using a message, and receives the long data directly from the application node without passing through the server A. As a result, the overhead caused by transferring the long data from a server to a server of another node is eliminated, and the load of the network is reduced.
摘要:
A computer system having a plurality of host computers and a storage system is provided which allows any one host computer to perform a global copy operation on any arbitrary or all storage areas in the storage system. To this end, storage areas provided by the disk devices are grouped into groups by allocating group numbers to a plurality of specified storage areas. The copy operation can be performed by specifying desired groups. Each of the groups is made up of sub-groups and the sub-groups are defined for each computer to assure a consistency of copy order of the sub-groups.
摘要:
Deduplicated backup data of a plurality of generations are aggregated and stored.A storage apparatus is connected via a network to a host system making a content write request and includes a storage unit providing one or more containers composed of a specified storage area, and a back up unit storing the content in the container in accordance with a backup request from the host system; and wherein the backup unit cuts out the content into one or more chunks, detects a duplicate chunk, which is a duplicate of a chunk stored in the container, from the cutout chunks, and additionally writes the chunk, other than the duplicate chunk, to the container where the duplicate chunk is stored.
摘要:
An optimum chunk cutout method is selected according to the type of content.A storage apparatus is a storage apparatus for storing content in a backup volume in response to a content storage request from a host system connected to the storage apparatus via a network and includes a chunk cutout unit for cutting out the content into one or more chunks and a duplication judgment unit for managing a duplicate state of the chunk or chunks which have been cut out by the chunk cutout unit; wherein the chunk cutout unit selects a method for cutting out the chunk based on content type identification information indicating a type of the content.
摘要:
A read/write replica of a file in a file server is updated through a communication line by a client. At the same time, a replica server detects the update data by using a virtual communication line between the client and a read-only replica of the file. The update data are used for refreshing the read-only replica. The refreshing process makes it unnecessary to send the update data from the file server to the replica server. Also, communication is made between the file server and the replica server to confirm validity of the update information taken in by the replica server to maintain equality of the files of the file server and the replica server. This can efficiently refresh the read-only replica of the replica server in a distributed data processing system without re-sending the update data.
摘要:
In a memory system having a storage device and a key storage keeping key data controlling an access to the storage device, there is disposed a key address translation structure for obtaining an address of an entry of the key storage based on an address of the storage device to which an access request is issued. As a result, when subdividing the storage device according to the key data setting unit, each subdivided area can be assigned with a variable size and a plurality of sizes are enabled to be specified for the key data setting units at the same time.
摘要:
Deduplicated backup data of a plurality of generations are aggregated and stored.A storage apparatus is connected via a network to a host system making a content write request and includes a storage unit providing one or more containers composed of a specified storage area, and a back up unit storing the content in the container in accordance with a backup request from the host system; and wherein the backup unit cuts out the content into one or more chunks, detects a duplicate chunk, which is a duplicate of a chunk stored in the container, from the cutout chunks, and additionally writes the chunk, other than the duplicate chunk, to the container where the duplicate chunk is stored.
摘要:
A computer connected to a plurality of computers, having: a storage area for storing data representative of properties of the plurality of computers relative to the computer; a data area for storing data to be used by the plurality of computers; and a program for determining a computer which holds data in the data area in accordance with a request for the data area. The computer can recover data lost from the computer from a proper area. Each of the properties of the plurality of computers is represented by a value determined by at least one of a computer fault occurrence frequency, a data transfer speed relative to the computer, an importance degree of data shared by processing programs of the plurality of computers. In accordance with the value in the storage area, the data in the data area,can be stored in multiples or in divisions.
摘要:
A method of controlling data which is stored into a common storage in a multiprocessor system having a common storage which can be accessed from a plurality of computers is disclosed. Particularly, a method of controlling shared data using an extended storage as a common storage is disclosed. The copy of the data of a part of or all of a shared data set in an input/output device and the copy data control information to control the copied data are stored into the common extended storage. In the case of accessing the data in the common extended storage, the mutual exclusion among a plurality of computers and the consistency control of the data are realized by referring to the control information stored in the common extended storage.
摘要:
An area boundary between a system common area and a job private area is set at any page boundary independently from a segment boundary, and for the segment (boundary segment) containing the area boundary, a page table is prepared for each virtual address space. Thus, virtual pages which are not used as the system common area in the bondary segment can be used by jobs as job private areas.The real page is fixedly allocated to the virtual page belonging to the system common area in the boundary segment. Thus, it is not necessary to simultaneously update page tables for the system common area. Those virtual pages may be subjects of dynamic allocation of the virtual storage.