Emc Corporation

Array Technology, LLC

Other

Nstor Corporation

Ambex Technologies, Inc.

David W. Gordon

Walter A. Hubis

David C. Stallmo

David M. Warner

William A. Brant

David C Stallmo

William A Brant

David W Gordon

Randy Kay Hall

Randy Hall

Michael Edward Nielson

Mark S Walker

Albert S Lui

Walter A Hubis

Gary Neben

Anthony D Andrews

Candace Brinkman

David M Warner

A RAID-compatible data storage system which allows incremental increases in storage capacity at a cost that is proportional to the increase in capacity. The system does not require changes to the host system. The control and interface functions previously performed by a single (or redundant) central data storage device controller are distributed among a number of modular control units (MCUs). Each MCU is preferably physically coupled to a data storage device to form a basic, low-cost integrated storage node. One of two bus ports interfaces an MCU with the host computer on a host bus, and the other bus port interfaces an MCU with one or more data storage devices coupled to the MCU by a data storage device bus. The serial interface ports provide a means by which each of the MCUs may communicate with each other MCU to facilitate the implementation of a memory array architecture. The entire data storage array may appear as a single device capable of responding to a single identification number on the host bus, or may appear as a number of independent devices. A controlling MCU receives a command and notifies the other MCUs that are involved in a read or write operation. Control of the host bus is transferred from one MCU to the next MCU in sequence so that the data is received by the host computer, or written to each data storage device, in the proper order.

Distributed storage array system having plurality of storage devices which each of devices including a modular control unit for exchanging configuration information over a communication link

A redundant array storage system that can be configured as a RAID 1, 3, 4, or 5 system, or any combination of these configurations. The invention comprises a configuration data structure for addressing a redundant array storage system, and a method for configuring a redundant array storage system during an initialization process. The redundant array storage system comprises a set of physical storage units which are accessible in terms of block numbers. The physical storage units are each configured as one or more logical storage units. Each logical storage unit is addressed in terms of a channel number, storage unit number, starting block number, offset number, and number of blocks to be transferred. Once logical storage units are defined, logical volumes are defined as one or more logical storage units, each logical volume having a depth characteristic. After the logical volumes are defined, redundancy groups are defined as one or more logical volumes. A redundancy level is specified for each redundancy group. The redundancy level may be none, one, or two. Logical volumes are addressed by a host CPU by volume number, initial block number, and number of blocks to be transferred. The host CPU also specifies a READ or WRITE operation. The specified volume number, initial block number, and number of blocks to be transferred are then translated into a corresponding channel number, storage unit number, starting block number, offset number, and number of blocks to be transferred. With the present invention, it is possible for a logical volume to span across physical storage units ('vertical partitioning'), comprise only a portion of each such physical storage unit ('horizontal partitioning'), and have definable depth and redundancy characteristics.

Logical partitioning of a redundant array storage system

A Redundant Array of Independent Disks (RAID) system that is adaptable to host Input/Output traffic, wherein the RAID configuration is hidden from the host computer, and wherein the system uses various methods for striping and organizing data across the array when disks are added or removed. The system dynamically determines the RAID configuration used to store host data. To maximize response time and avoid a write penalty, small write operations are mapped into RAID 1 configurations, and medium and large write operations are mapped into RAID 3 configurations. The data are migrated into RAID 5 configurations as a background operation, to minimize the disk space lost. The system hides configuration changes necessary for the addition and/or removal of disks to/from the disk array by unstriping data, adding or deleting (removing) a disk and then restriping data over all the disks. The methods may include steps of dividing data blocks into a plurality of square portions, and exchanging data in sets of blocks of each of the square portions, with the exchanging step including locating a diagonal set of blocks within a selected square portion, and exchanging all sets of blocks equidistant from the diagonal, on opposite sides of the diagonal, and in a line perpendicular to the diagonal. While these changes are in progress, the disk array remains on-line and most host data is available for access and modification.

Method for organizing storage devices of unequal storage capacity and

A fault-tolerant storage device array using a copyback cache storage unit for temporary storage. When a Write occurs to the RAID system, the data is immediately written to the first available location in the copyback cache storage unit. Upon completion of the Write to the copyback cache storage unit, the host CPU is immediately informed that the Write was successful. Thereafter, further storage unit accesses by the CPU can continue without waiting for an error-correction block update for the data just written. In a first embodiment of the invention, Read-Modify-Write operations are performed during idle time. In a second embodiment of the invention, normal Read-Modify-Write operation by the RAID system controller continue use Write data in the controller's buffer memory. In a third embodiment, at least two controllers, each associated with one copyback cache storage unit, copy Write data from controller buffers to the associated copyback cache storage unit. If a copyback cache storage unit fails, more than one controller share a single copyback storage unit. In a fourth embodiment, Write data is copied from a controller buffer to a reserved area of each storage unit comprising the array.

Storage device array architecture with copyback cache

Storage device array and methods for striping and unstriping data and

A RAID-compatible data storage system which allows incremental increases in storage capacity at a cost that is proportional to the increase in capacity. The system does not require changes to the host system. The control and interface functions previously performed by a single (or redundant) central data storage device controller are distributed among a number of modular control units (MCUs). Each MCU is preferably physically coupled to a data storage device to form a basic, low-cost integrated storage node. One of two bus ports interfaces an MCU with the host computer on a host bus, and the other bus port interfaces an MCU with one or more data storage devices coupled to the MCU by a data storage device bus. The serial interface ports provide a means by which each of the MCUs may communicate with each other MCU to facilitate the implementation of a memory array architecture. The entire data storage array may appear as a single device capable of responding to a single identification number on the host bus, or may appear as a number of independent device. A controlling MCU receives a command and notifies the other MCUs that are involved in a read or write operation. Control of the host bus is transferred from one MCU to the next MCU in sequence so that the data is received by the host computer, or written to each data storage device, in the proper order.

Distributed disk array architecture

Distributed storage array system having a plurality of modular control

A RAID disk array that is adaptable to host I/O traffic, wherein the RAID configuration is hidden from the host computer. The system dynamically determines the RAID configuration used to store host data to maximize response time performance and minimize the loss of disk space used for data protection. To maximize response time and avoid a write penalty, small write operations are mapped into RAID 1 configurations, and medium and large write operations are mapped into RAID 3 configurations. These segments are migrated into RAID 5 configurations as a background operation, to minimize the disk space lost. The system hides configuration changes necessary for the addition and/or deletion of disks to the disk array. While these changes are in progress, the disk array remains on-line and all host data is available for access and modification.

Method and apparatus for improving performance in a reduntant array of

A method for on-line restoration of redundancy information in a redundant array storage system. The invention provides alternative methods of restoring valid data to a storage unit after a Write failure caused by a temporary storage unit fault. In the first preferred method, a valid redundancy block is generated for the corresponding data blocks on all storage units. Resubmitting the interrupted Write operation causes the old (and potentially corrupted) data block to be 'subtracted' out of the re-computed redundancy block. The uncorrupted new data block is written over the old data block, and is 'added' into the re-computed redundancy block to create a new, corrected redundancy block. The new, corrected redundancy block is written to the appropriate storage unit. In the second preferred method, a new redundancy block is generated from all valid data blocks and the new data block. The new redundancy block and the new data block are then written to the appropriate storage units. In both cases, the entire method is done on-line, with insignificant interruption of normal operation of the redundant array system, and without requiring added processing during normal operation.

On-line restoration of redundancy information in a redundant array system

An array controller including a DATA-RAM and a SHADOW-RAM. Both the DATA-RAM and the SHADOW-RAM are coupled to a first and second memory interface. Each memory interface has the ability to independently communicate the contents of the SHADOW-RAM over a controller-controller data link to at least one other similar array controller. The memory interface also interfaces the DATA-RAM and the SHADOW-RAM to a CPU, the data storage units of the RAID system, and the controller processor. Write data received from the CPU is stored in the two independent memories in order to ensure that pending Write data (i.e., Write data that has not yet been written to the RAID system, including any copyback cache device) will not be lost. In addition, the two memory interfaces provide redundant access routes which allow Write data to be retrieved by another array controller if the controller processor fails. A backup power source is provided to ensure that power will be available to at least one of the two memories such that the data that has been received within the controller will always be accessible. Accordingly, since the data will be accessible, even if of a failure of any single component or power source occurs, the controller may acknowledge the write operation requested by the CPU as soon as data is successfully written to both the DATA-RAM and the SHADOW-RAM.

Method and apparatus for fault tolerant fast writes through buffer

A redundant array storage system including storage units divided into two logical arrays. The redundant array storage system further includes a plurality of array control units which are all fully utilized to control data transfers between the logical arrays and a central processing unit, each controller being capable of taking over the task of a failed controller. In normal operation, each redundant array controller may only access data stored in a logical array assigned to that controller. If the other redundant array controller fails, the remaining controller may access the data stored in the logical array assigned to the failed controller only through a secondary control process that is independent from the primary control process of the remaining controller. Thus, the invention prevents parity data associated with user data placed in storage from being corrupted by attempts of two or more array control units to access the same redundancy group of data concurrently.

Multiple controller sharing in a redundant storage array

A redundant array storage system that can be configured as a RAID 1, 3, 4, or 5 system, or any combination of these configurations. The invention includes a configuration data structure for addressing a redundant array storage system, and a method for configuring a redundant array storage system during an initialization process. The redundant array storage system includes a set of physical storage units which are accessible in terms of block numbers. The physical storage units are each configured as one or more logical storage units. Each logical storage unit is addressed in terms of a channel number, storage unit number, starting block number, offset number, and number of blocks to be transferred. Once logical storage units are defined, logical volumes are defined as one or more logical storage units, each logical volume having a depth characteristic. After the logical volumes are defined, redundancy groups are defined as one or more logical volumes. A redundancy level is specified for each redundancy group. The redundancy level may be none, one, or two. Logical volumes are addressed by a host CPU by volume number, initial block number, and number of blocks to be transferred. The host CPU also specifies a READ or WRITE operation. The specified volume number, initial block number, and number of blocks to be transferred are then translated into a corresponding channel number, storage unit number, starting block number, offset number, and number of blocks to be transferred. With the present invention, it is possible for a logical volume to span across physical storage units ('vertical partitioning'), comprise only a portion of each such physical storage unit ('horizontal partitioning'), and have definable depth and redundancy characteristics.

Growing community of inventors

Boulder, CO, United States of America

David C Stallmo