Hitachi, Ltd.

Yutaka Enko

Yoji Nakatani

Masaaki Iwasaki

Takahiro Nakano

Toshiaki Arai

Tetsuya Mochida

Tomoki Sekiguchi

Tsuyoshi Watanabe

Tsutomu Noda

Masami Hiramatsu

Satoshi Oshima

Naohiro Fujii

Reki Yamamoto

Naofumi Shouji

The present invention relates to a migration of a file from a path denotative NAS server to an ID denotative NAS server. In a file management device a migration-use path denotative NAS access section obtains a file deposit location from the path denotative NAS server, selects a migration target file based on the deposit location, and sends a read request with the deposit location of the migration target file to the path denotative NAS server, so as to obtain the migration target file. A migration-use denotative NAS access section generates a write request with the migration target file, sends it to the ID denotative NAS servers, and writes the file therein. In addition, the migration-use ID denotative NAS access section registers a GUID received from the ID denotative NAS servers in a file name/GUID management table, together with the file name of the migration target file.

File storage service system, file management device, file management method, ID denotative NAS server and file reading method

The present invention relates to a migration of a file from a path denotative NAS server to an ID denotative NAS server. In a file management device according to one embodiment, a migration-use path denotative NAS access section obtains a file deposit location from the path denotative NAS server, selects a migration target file based on the deposit location, and sends a read request with the deposit location of the migration target file to the path denotative NAS server, so as to obtain the migration target file. A migration-use denotative NAS access sectiongenerates a write request with the migration target file, sends it to the ID denotative NAS servers, and writes the file therein. In addition, the migration-use ID denotative NAS access section registers a GUID received from the ID denotative NAS servers in a file name/GUID management table, together with the file name of the migration target file.

A meta data server, a user nodeand a storage deviceare connected together via a Fibre channeland a LANto configure a NAS over SAN system. The user nodesends out a request to access the data in each file. The file is stored in the storage devicein blocks of default size. The meta data serverinforms the user nodeof the block number where the file to be accessed is stored based on meta data in which relationship between a file and a block is recorded. Prior to this processing, the meta data serverinstructs the storage deviceto cache the data in advance. This application enables the storage deviceto provide the data promptly in response to read request from the user node

Data prefetch in storage device

In order to access a distributed file system (DFS) of the present invention using a conventional protocol such as the one for an NFS or a CIFS without making a modification on a client side, a gateway unit for receiving a conventional protocol and performing processing in conformity the protocol is provided for a DFS server. The gateway unit emulates a directory structure in a file system such as the NFS or CIFS. When the DFS is a write-once read-many file system, update processing is converted into processing for creating a new generation file, and reference processing is converted into access to a latest generation file in a generation-managed file group. The gateway unit then accesses a DFS file via a DFS processing unit.

Method for accessing distributed file system

In a network system in which clients and file servers are connected by networks and a client makes access to a file stored in a disk storage device of a file server via the networks, accessibility to a file of a file server distant from the client tends to be deteriorated. In order to resolve the problem, the file server having the file searches for an appropriate file server in the proximity of the client and transfers the file to the appropriate file server (migration target server). Thereafter, the client makes access to the migration target server for the file.

File migration device

A computer system is structured, including a plurality of computers sharing a storage apparatus. The computers each include a control section for determining an access priority based on an attribute of a client computer forwarding an access request. In the control section, the result derived by adding up the priorities is compared with the total value of the priorities calculated by other computers for assignment of an input/output bandwidth. In accordance with the priority assignment, access execution is performed in order of priority. With such a structure, access congestions can be favorably avoided through efficient bandwidth control, and bandwidth assignment can be performed on a server basis in accordance with processes executed by the client computers.

Computer system and bandwidth control method for the same

Providing a network protocol processing device by which a server connected to a network logically parted on a user basis can offload TCP protocol processing. In a server having a network protocol processing function including: a host processor that processes user requests received through a network; and TOE equipment including a network port that is connected with the host processor, and connected with the network, a processor that performs TCP/IP protocol processing, and a memory that holds TCP connection identification information, information indicating a source IP address, a destination IP address, a source port number, and a destination port number, and frame information including VLAN-ID defined by IEEE802.1Q are used as TCP connection identification information.

Network protocol processing device

The parallelism of execution of processes in a memory shared multi-processor is enhanced. Each of a plurality of processors is provided with a run queue for holding executable processes, and a watch process for each processor to watch the execution status of its own to notify the rebuilding of the run queue to other processor and a rebuild step for rebuilding the run queue in response to the notice are stored in a shared memory as a program.

Growing community of inventors

Sagamihara, Japan

Yutaka Enko