Hitachi, Ltd.

Other

Hitachi Computer Peripherals Co., Ltd.

Kenta Shiga

Keisei Fujiwara

Naoko Iwami

Atsuya Kumagai

Daiki Nakatsuka

Toshihiko Murakami

Naoto Matsunami

Yoshinori Ohira

Tetsuya Shirogane

Akira Nishimoto

Norihiko Kawakami

Keiko Tanigawa

Yoshiaki Eguchi

Naoki Watanabe

Shoji Kodama

Kazuyoshi Serizawa

Satoshi Kikuchi

Kazuma Yumoto

Toshio Otani

Wujuan Lin

Junji Fukuzawa

Yoko Hirashima

Azusa Sudo

Toru Hoshi

Akira Yamamoto

Junji Ogawa

Nobuhiro Maki

Koji Watanabe

Yusuke Nonaka

Yasuo Watanabe

Yoichi Mizuno

Masakazu Mitsuhashi

Takashi Amano

Hirokazu Ikeda

Akihiko Araki

Atsushi Ueoka

Masashi Suzuki

Satoshi Kiyoto

Kaushik Triyambaka Mysur

Hitoshi Yui

Tooru Tanaka

Lin Wujuan

Toru Ohashi

First type metadata is associated with unstructured data included in an unstructured data source. A data processing system performs an extraction process. This extraction process includes: (a) creating, for each of a plurality of selected pieces of unstructured data in the unstructured data source, second type metadata, which is metadata including content information representing one or more content attributes of the piece of unstructured data; and (b) associating the created second type metadata with the first type metadata of the piece of unstructured data.

Data processing system and data processing method

A computer system with a plurality of storage systems connected to each other via a network, each storage system including a virtual machine whose data is stored in hierarchized storage areas. When a virtual machine of a first storage system is migrated from the first storage system to a second storage system, the second storage system stores data of the virtual machine of the first storage system as well as data of its own virtual machine, in the hierarchized storage areas in the second storage system.

Computer system and control method for the same

An aspect of the invention is directed to a plurality of MDSs (metadata servers) in a distributed storage system which includes data servers storing file contents, each MDS having a processor and a memory and storing file system metadata. Directories of a file system namespace are distributed to the MDSs through referral directories referring to real directories using hash value of inode number of each of the referral directories. During a process to create a directory in the file system namespace, a first MDS dynamically creates a referral directory in the first MDS, and creates a real directory in a second MDS, the referral directory in the first MDS referring to the real directory in the second MDS by maintaining location information of the real directory in the second MDS, the real directory in the second MDS containing the file system metadata for the directory in the file system namespace.

Directory-level referral method for parallel NFS with multiple metadata servers

Proposed are a storage apparatus and its control method capable of restoring data of a virtual page to an appropriate device hierarchy according to its data value. This storage apparatus stores, in a backup-destination storage medium, data stored in the respective physical pages allocated with the virtual volume and management information of the virtual volume including information concerning the device hierarchy to which the respective physical pages belong according to a backup command of the virtual volume, and allocates a physical page of the same device hierarchy as the device hierarchy of the physical page that was allocated before the backup to each of the restoration-target virtual pages of the virtual volume, respectively, and writes corresponding data, which was read from the storage medium, into each of the allocated physical pages according to a restoration command of the virtual volume and based on the management information of the virtual volume stored in the storage medium.

Storage apparatus and its control method

The present invention provides a storage subsystem accessing a storage resource efficiently to thereby enhance throughput. The storage subsystem adopts a configuration where respective processor packages have ownership of the respective LDEVs, wherein an I/O command from a host specifying an address extending over a division position within a single LDEV is divided, and the I/O processing is executed in a shared manner by multiple processor packages. The I/O processing is executed by allocating a processor package that processes a specific storage area with priority.

Storage subsystem and method for controlling the storage subsystem

A pool is formed based on a plurality of storage devices. This pool is constituted by a plurality of real pages. Real pages of different lengths are included in this plurality of real pages. Among a plurality of virtual pages which make up a virtual volume, a controller compresses a write data element for a write destination virtual page, selects a real page of a real page length based on the data length of a data unit including the compressed write data element, and allocates the selected real page to the write destination virtual page.

Storage system and storage control method that compress and store data elements

The present invention aims at improving the performance of a compression function in a storage system, and solves the prior art problem of having to decompress a whole compression unit even if a read request or a write request targets only a portion smaller than the compression unit, causing increase of overhead of decompression processing and elongation of processing time, and deteriorating performance. The present invention prevents unnecessary decompression processing and reduces the overhead of processing by suppressing the range of decompression processing to a minimum portion according to the relationship between the read/write request range and the compression unit.

Method for decompressing data in storage system for write requests that cross compressed data boundaries

A distributed storage system employs a Key-Value Store to dynamically change a table layout scheme based on user access patterns. The system can be used as a basic architecture to construct a distributed Key-Value Store to support both row-oriented and column-oriented table layout schemes, by using DHT (Distributed Hash Table) technology for high scalability.

Data replication and failure recovery method for distributed key-value store

A first storage system includes a plurality of first storage devices and is coupled to a computer. A second storage system includes a plurality of second storage devices and is coupled to the first storage system. A first controller provides a thin provisioning logical volume (LU) to the computer. A second controller provides an external thin provisioning LU to the first storage system. The first controller provides pool areas associated with the thin provisioning LU, including a first pool area mapped to the external thin provisioning LU, and allocates the first pool area to a first region in the thin provisioning LU to store a write data to the first region in the thin provisioning LU. The second controller allocates at least one of a plurality of pool areas to store the write data to the first region in the thin provisioning LU.

Allocating storage to a thin provisioning logical volume

In a system of local DHT overlays, each has KVS nodes, including one super node. The super nodes organize a global DHT overlay. Each super node maintains Bloom filters of keys in its local DHT overlay for all key ranges. To obtain data corresponding to a key from other local DHT overlays, a super node sends a request to a node which is responsible for the key range hashed from the specified key. The responsible node determines local DHT overlays which may have data corresponding to the key according to Bloom filters registered in the super nodes of the local DHT overlays, updated and converted from counting filters. Requests are sent to the super nodes of the local DHT overlays identified by the responsible node. Thus, requests are not needlessly sent to super nodes of local DHT overlays which do not have data corresponding to the key, thereby reducing traffic.

Traffic reduction method for distributed key-value store

Storage system for a virtual volume across a plurality of storages

There is provided a storage subsystem having a virtual volume and a page volume which has a page physical area allocated to the virtual volume. The storage subsystem divides an address space of the virtual volume into a plurality of pages, classifies each of the pages into one of a plurality of states including at least a first state and a second state, and further divide a page which is classified into the second state into a plurality of segments to managed the page classified into the second state. The first state is a state in which a page physical area is allocated to the page from the page volume, and the write data is stored in the page physical area. The second state is a state in which the predetermined pattern data and the segment are managed, in the memory, by correlating with each other.

Storage subsystem

A storage controller comprises a cache storage used as a cache of an external storage and a control processor coupled to the cache storage. The control processor comprises an internal access function and an external access function. The internal access function transmits a read command to the cache storage. The cache storage determines whether the read target data complying with the read command is stored in a physical storage device or not. If the result of the determination is negative, the external access function is executed while, if the result of the determination is positive, the external access function is not executed because the cache storage transmits the read target data to the internal access function without issuing a request to the external access function.

Storage controller coupled to storage apparatus

To enhance capacity expansion property of a storage system for providing thin provisioning logical volumes, a computer system includes a first storage system and a second storage system for providing a second volume to the first storage system. The first storage system is configured to define at least one first volume, define a first storage pool including the defined first volume and the provided second volume, and provide to the host computer a first thin provisioning volume that is recognized as a volume having a capacity equal to or larger than that of storage areas assigned from the first storage pool. The first thin provisioning volume is thus assigned with both storage areas of the first volume and storage areas of the second volume, which are included in the first storage pool.

Management method for a virtual volume across a plurality of storages

Growing community of inventors

Yokohama, Japan

Kenta Shiga