摘要:
A network-based storage system comprises one or more block-level storage servers that connect to, and provide disk storage for, one or more host computers. In one embodiment, the system is capable of subdividing the storage space of an array of disk drives into multiple storage partitions, and allocating the partitions to host computers on a network. A storage partition allocated to a particular host computer may appear as local disk drive storage to user-level processes running on the host computer.
摘要:
A network-based storage system comprises one or more block-level storage servers that connect to, and provide disk storage for, one or more host computers (“hosts”) over logical network connections (preferably TCP/IP sockets). In one embodiment, each host can maintain one or more socket connections to each storage server, over which multiple I/O operations may be performed concurrently in a non-blocking manner. The physical storage of a storage server may optionally be divided into multiple partitions, each of which may be independently assigned to a particular host or to a group of hosts. When a host initially connects to a storage server in one embodiment, the storage server initially authenticates the host, and then notifies the host of the ports that may be used to establish data connections and of the partitions assigned to that host.
摘要:
A network-based storage system comprises one or more block-level storage servers that connect to, and provide disk storage for, one or more host computers (“hosts”) over logical network connections (preferably TCP/IP sockets). In one embodiment, each host can maintain one or more socket connections to each storage server, over which multiple I/O operations may be performed concurrently in a non-blocking manner. The physical storage of a storage server may optionally be divided into multiple partitions, each of which may be independently assigned to a particular host or to a group of hosts. Host driver software presents these partitions to user-level processes as one or more local disk drives. When a host initially connects to a storage server in one embodiment, the storage server initially authenticates the host, and then notifies the host of the ports that may be used to establish data connections and of the partitions assigned to that host.
摘要:
The present invention is related to methods and systems for improving the read performance of a drive array, such as a RAID 5 array, having an “n” number of drives so that the read performance is greater than (n−1) times that of an independent single drive. Formatting information for each of at least three drives is received. A first parity block size for use with corresponding first zones of the at least three drives is selected. A second parity block size different than the first parity block size is selected for use with corresponding second zones of the at least three drives. The first and the second parity block sizes are selected to increase the read performance from the drive array based on at least a portion of the formatting information.
摘要:
The present invention is related to methods and systems for providing different stripe sizes for different zones for at least a first of a plurality of mirrored drives to improve data rates. The first drive has a plurality of zones. In one embodiment, a first stripe size is selected for a first zone, and a second stripe size is selected for a second zone. The said second stripe size is different than said first stripe size.
摘要:
The present invention is related to methods and systems for accessing multimedia data stored on a disk array to ensure that the transfer rate does not fall below a selected minimum transfer rate independent of where the multimedia data is stored on the disk array. In one embodiment, the sizes of blocks accessed are varied and the direction of access is selected to enhance read performance. A plurality of blocks are read whose sizes vary linearly as data is being read from a first disk in a backward direction, from an inner diameter side of the first disk, towards an outer diameter of the first disk. Blocks are read from a second disk in a forward direction, towards an inner diameter of the second disk.
摘要:
A network-based storage system comprises one or more block-level storage servers that connect to, and provide disk storage for, one or more host computers. In one embodiment, the system is capable of subdividing the storage space of an array of disk drives into multiple storage partitions, and allocating the partitions to host computers on a network. A storage partition allocated to a particular host computer may appear as local disk drive storage to user-level processes running on the host computer.
摘要:
A network-based storage system comprises one or more block-level storage servers that connect to, and provide storage for, one or more host computers over logical network connections, such as TCP/IP connections. In one embodiment, the block-level storage servers implement a protocol through which a storage server authenticates a host before permitting the host to access storage resources. Upon successful authentication, the storage server may also provide access information to the host.
摘要:
The present invention is related to methods and systems for reducing head movements during accesses to drives within a drive array by dynamically selecting which one of a first drive and a second drive is to be used to read data stored within a first logical address range. A first set of data is stored in both the first drive and the second drive. The logical block addresses accessed by at least a portion of read operations are monitored. The first drive is designated as a data source for data stored on both the first drive and the second drive in the first logical address range. The second drive is designated as a data source for data stored on both the first drive and the second drive outside the first logical address range. The designation of drives is performed at least partly in response to the monitoring act.