Multi-actuator drive providing replication using multiple volumes
阅读说明:本技术 使用多个卷提供复制的多致动器驱动器 (Multi-actuator drive providing replication using multiple volumes ) 是由 V·南俊达斯瓦米 于 2019-08-13 设计创作,主要内容包括:本申请公开了使用多个卷提供复制的多致动器驱动器。第一驱动器卷在驱动器外壳内的盘上形成。由第一头从第一驱动器卷读取以及向第一驱动器卷写入,该第一头由第一致动器移动。在一个或多个盘上形成第二驱动器卷。由第二头从第二卷读取以及向第二卷写入,该第二头经由驱动器外壳内的第二致动器移动。第二致动器是与第一致动器分开的并且独立于第一致动器。第一驱动器卷的数据被复制到第二驱动器卷上。在第二驱动器卷而不是第一卷上执行后台验证操作。(The present application discloses a multi-actuator drive that provides replication using multiple volumes. The first drive roll is formed on a disk within a drive enclosure. The method includes reading from and writing to a first drive volume by a first head, the first head being moved by a first actuator. A second drive volume is formed on the one or more disks. Reading from and writing to the second volume is performed by a second head that is moved via a second actuator within the drive housing. The second actuator is separate and independent from the first actuator. Data of the first drive volume is copied onto the second drive volume. A background authentication operation is performed on the second drive volume instead of the first volume.)
1. A data storage drive comprising:
an interface circuit configured to communicate with:
a first actuator and a second actuator enclosed within the independently operated driver; and
a first read/write head and a second read/write head moved over one or more disks by the first actuator and the second actuator, respectively; and
a controller coupled to the interface circuit and operable to:
forming a first drive volume of the one or more disks, read from and write to the first drive volume by the first read/write head;
forming a second drive volume for the one or more disks, read from and written to the second drive volume by the second read/write head, the second volume replicating data of the first drive volume; and
performing a background authentication operation on the second drive volume instead of the first drive volume.
2. The data storage drive of claim 1, wherein performing the background authentication operation on the second drive volume instead of the first volume minimizes a performance impact on the first volume.
3. The data storage drive of claim 1, the controller further operable to: testing similar errors on the first volume using errors found on the second volume from the background verification operations.
4. The data storage drive of claim 1, wherein the second volume is a mirror image of the first volume.
5. The data storage drive of claim 1, wherein the second volume comprises data of the first volume stored in a reduced format.
6. A method, comprising:
forming a first drive volume of one or more disks within a drive enclosure, reading from and writing to the first drive volume by a first read/write head, the first read/write head moved by a first actuator within the drive enclosure;
a second drive roll forming the one or more disks, read from and write to the second drive roll by a second read/write head, the second read/write head moved via a second actuator within the drive enclosure, the second actuator separate and independent from the first actuator;
copying data of the first drive volume onto the second drive volume; and
performing a background authentication operation on the second drive volume instead of the first volume.
7. The method of claim 6, further comprising: testing similar errors on the first volume using errors found on the second volume from the background verification operations.
8. The method of claim 6, further comprising: reading from the first volume in parallel with the first head and the second head to increase throughput of host data requests.
9. A system, comprising:
one or more disks in the drive enclosure;
first and second actuators that independently move respective first and second heads over the one or more disks; and
a controller coupled to the first and second actuators and the first and second heads, and operable to:
defining a first volume on a first portion of the one or more disks, reading from and writing to the first portion by the first read/write head;
defining a second volume on a second portion of the one or more disks, read from and written to the second portion by the second read/write head;
in response to writing data to the first volume, copying the data to the second drive volume; and
performing a background authentication operation on the second volume instead of the first volume.
10. The system of claim 9, wherein the first actuator and the second actuator move about separate pivots, and wherein at least a portion of the first roll and at least a portion of the second roll are located on a same surface of the one or more disks.
Disclosure of Invention
The present disclosure relates to a multi-actuator for providing replication using multiple volumes. In one embodiment, a first drive volume of one or more disks within a drive enclosure is formed. The first read/write head is moved by a first actuator within the drive enclosure to read from and write to the first drive volume. A second drive roll forming one or more disks is read from and written to the second drive roll by a second read/write head that is moved via a second actuator within the drive enclosure. The second actuator is separate and independent from the first actuator. Data of the first drive volume is copied onto the second drive volume. A background authentication operation is performed on the second drive volume instead of the first volume.
These and other features and aspects of the various embodiments can be understood from the following detailed discussion and the accompanying drawings.
Drawings
The following discussion makes reference to the following drawings, wherein like reference numerals may be used to identify similar/identical components in the various figures.
FIG. 1 is a diagram of a plurality of actuator drivers, according to an example embodiment;
FIG. 2 is a diagram illustrating a drive roll configuration utilizing a split actuator, according to an example embodiment;
FIG. 3 is a diagram illustrating a drive volume configuration on a single disk surface in accordance with an example embodiment;
FIG. 4 is a diagram illustrating a drive volume configuration on a single disk surface in accordance with another example embodiment;
FIG. 5 is a diagram illustrating a logical configuration of a drive volume according to an example embodiment;
FIG. 6 is a block diagram of a system and apparatus according to an example embodiment; and
fig. 7 is a flow chart of a method according to an example embodiment.
Detailed Description
The present disclosure relates generally to data storage devices that utilize magnetic storage media, such as Hard Disk Drives (HDDs). The additional HDD features described herein (often described as a "parallelism" architecture) are considered a way to increase HDD performance metrics such as data throughput and latency. Typically, a parallelism architecture operates multiple read/write heads in parallel. Such parallelism may increase the rate of input/output operations (IOPS) and thereby speed up certain operations. For example, data read from two heads may be combined together into a single stream, thereby doubling the throughput rate of data sent to the host. In other examples, different headers may serve different read or write requests simultaneously, thereby reducing overall latency, for example, for multiple simultaneous random data access requests.
In the embodiments described below, a hard disk drive includes multiple heads driven by different actuators that can read from or write to one or more disks simultaneously. This may include separate and independent reads/writes, such as headers that service different read/write requests. This may also include separate and related reads/writes, for example, where multiple portions of a single data stream are processed simultaneously by different heads. In either case, the head and the actuator themselves are operating independently, although in the latter case coordinated. The parallelism architecture can be extended to other components operating in the HDD, including a system controller, a servo controller, read/write channels, a host interface, cache, and the like.
While parallelism is generally considered a technique for increasing data throughput, it can also be used to increase reliability while minimizing performance impact, as described below. For example, a drive may lose data due to errors that affect the read/write head and/or the recording medium. In storage arrays that use redundancy, such as Redundant Arrays of Independent Disks (RAIDs), a plurality of disk drives are assembled to form a logical volume, with redundant data stored on some of the disks in some RAID configurations. The redundant data may be used to reconstruct data lost due to media/head errors. Due to the very large storage capacity of modern drives, the time taken to reconstruct the data can be long, hours or even days. Thus, it is possible that another drive of the array fails while disk reconstruction is occurring. Depending on the type of RAID volume, this may still result in data loss.
One method of mitigating data loss is data scrubbing (scrub) of the disc. Typically, this involves reading previously written data and checking for errors. This can be used to find and repair errors, or used as a predictor for future errors, such as a high error rate when reading data. In either case, the lost (unrecoverable) data may be recovered, for example, from the RAID redundancy data, and overwritten in the same or a different location. Typically, the cleaning at the drive level or disk group level is slow and runs at a low priority. These cleans are typically disabled by the user because it degrades system performance and results in cost.
In the embodiments described below, the drives have parallelism features that variously allow the formation of replicated volumes within multiple separate drives. Some parallelism features, such as dual actuators, may allow the drive to operate as two independent drive portions. These drive portions may be used to provide redundancy similar to RAID, but using only a single drive. One actuator (and the portion of the disk associated with that actuator) may be used as the primary drive, while the other actuator (and the disk portion associated with that actuator) is used for redundancy. Data may be replicated between the two drive portions either entirely (e.g., each byte of data on the primary drive is written on the backup) or by using a parity or compression scheme (e.g., a reduced form of each byte of data on the primary drive is written on the backup). The drive portions are also referred to herein as "volumes," but the two drive portions need not be presented to the host as volumes. For example, in some configurations, a host computer may only have access to a portion of the master disk via a host interface, which the host may use as an original partition that may be divided into one or more volumes. Invisible to the host, the drive controller may internally manage the backup disk portion and copy the data of the primary drive onto the backup portion. Backup data may include partition metadata, file system metadata, user data, parity data, and the like.
In FIG. 1, an apparatus 100 (e.g., a data storage drive) having parallelism features is illustrated, according to an example embodiment. The
The
In another embodiment, the
In the example shown in fig. 1, more than one
One or
As described above, the
In one embodiment, the
In fig. 3, a multiple actuator driver configuration according to another example embodiment is illustrated. In this example,
The radially defined
Note that in the arrangement shown in fig. 3, the
The arrangement shown in fig. 3 may be repeated for multiple disk surfaces of one or more disks, each disk having its own arm and head driven by the same actuator that drives the illustrated
Note that in this example,
In some arrangements, it may be preferable for two or more radially defined regions on the disk surface to have approximately the same performance. A zone arrangement according to an example embodiment, which may mitigate performance differences, is shown in fig. 4. The
In other arrangements, a drive with two or more actuators having different axes of rotation may still use a full disk surface for the drive roll. Referring again to fig. 3, the
In fig. 5, a block diagram illustrates a logical view of a
The
If the
The data stored by the
The
Note that the
During operation, integrity verification operations by which the
At least some of the background authentication operations performed on the
In some cases, errors found on
In FIG. 6, a block diagram illustrates a
The read/
The read/
The
An optional
In fig. 7, a flow chart shows a method according to an example embodiment. The method involves: a first drive volume of one or more disks of a data storage drive is formed 700. For example, a first portion of the disc may be reserved and defined for a first volume. The first drive volume is accessed 701 (e.g., read from and written to) via a first read/write head that is moved by a first actuator. A second drive volume of the one or more disks is formed 702, for example by reserving and defining a second portion of the disk for the second volume. The second volume is accessed 703 via a second read/write head, which is moved via a second actuator. The second actuator is separate from and independent of the first actuator. The first and second actuators and the one or more disks are enclosed within a common drive enclosure, such as an enclosure that conforms to a hard disk drive form factor.
Data for the first drive volume is copied 704 (e.g., in real-time or via a background process) onto the second drive volume. Background authentication operations are performed 705 on the second drive volume instead of the first volume. This may include, for example: operations are performed on the second volume at a higher frequency than the first volume. In some configurations, this may include: on the second volume, certain operations are performed periodically in the background, with the same operations being performed on the first volume except for elapsed time, only in response to an error or other triggering event.
The various embodiments described above may be implemented using circuits, firmware, and/or software modules that interact to provide specific results. One skilled in the relevant art can readily use knowledge known in the art to implement such described functionality at a modular level or as a whole. For example, the flow diagrams and control diagrams illustrated herein may be used to create computer readable instructions/code for execution by a processor. Such instructions may be stored on a non-transitory computer readable medium and transferred to a processor for execution, as is known in the art. The structures and processes shown above are merely representative examples of embodiments that may be used to provide the functionality described above.
The foregoing description of the example embodiments has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the embodiments to the precise form disclosed. Many modifications and variations are possible in light of the above teaching. Any or all of the features of the disclosed embodiments may be used alone or in any combination and are not intended to be limiting but merely illustrative. It is intended that the scope of the invention be limited not by this detailed description, but rather by the claims appended hereto.
The following are further examples of the present application:
example 1.
A data storage drive comprising:
an interface circuit configured to communicate with:
a first actuator and a second actuator enclosed within the independently operated driver; and
a first read/write head and a second read/write head moved over one or more disks by the first actuator and the second actuator, respectively; and
a controller coupled to the interface circuit and operable to:
forming a first drive volume of the one or more disks, read from and write to the first drive volume by the first read/write head;
forming a second drive volume for the one or more disks, read from and written to the second drive volume by the second read/write head, the second volume replicating data of the first drive volume; and
performing a background authentication operation on the second drive volume instead of the first drive volume.
Example 2. the data storage drive of example 1, wherein performing the background verification operation on the second drive volume instead of the first volume minimizes a performance impact on the first volume.
Example 3. the data storage drive of example 1, the controller further operable to: testing similar errors on the first volume using errors found on the second volume from the background verification operations.
Example 4. the data storage drive of example 1, wherein the second volume is a mirror image of the first volume.
Example 5 the data storage drive of example 1, wherein the second volume comprises data of the first volume stored in a reduced format.
Example 6 the data storage drive of example 1, wherein the second volume comprises parity data calculated based on data from different sectors of the first volume.
Example 7 the data storage drive of example 1, wherein the background authentication operation comprises: the data written to the second drive volume is read and an unrecoverable error is determined.
Example 8 the data storage drive of example 1, wherein the background authentication operation comprises: the data written to the second drive volume is read and a recoverable error is determined.
Example 9 the data storage drive of example 1, wherein the first actuator and the second actuator move about separate pivots and at least a portion of the first roll and at least a portion of the second roll are located on two or more radially defined areas positioned on a same surface of the one or more disks such that the second head can read from the first roll in parallel with the first head to increase throughput of host data requests.
Example 10 the data storage drive of example 9, wherein the two or more radially defined regions comprise four or more regions divided into two groups, the regions of each of the two groups being staggered and distributed from an inner diameter of the one or more disks to an outer diameter of the one or more disks.
Example 11 the data storage drive of example 1, wherein the first actuator and the second actuator move about a common pivot axis, and at least a portion of the first volume and at least a portion of the second volume are located on different surfaces throughout the one or more disks.
Example 12. a method, comprising:
forming a first drive volume of one or more disks within a drive enclosure, reading from and writing to the first drive volume by a first read/write head, the first read/write head moved by a first actuator within the drive enclosure;
a second drive roll forming the one or more disks, read from and write to the second drive roll by a second read/write head, the second read/write head moved via a second actuator within the drive enclosure, the second actuator separate and independent from the first actuator;
copying data of the first drive volume onto the second drive volume; and
performing a background authentication operation on the second drive volume instead of the first volume.
Example 13. the method of example 12, further comprising: testing similar errors on the first volume using errors found on the second volume from the background verification operations.
Example 14. the method of example 12, further comprising: reading from the first volume in parallel with the first head and the second head to increase throughput of host data requests.
Example 15. a system, comprising:
one or more disks in the drive enclosure;
first and second actuators that independently move respective first and second heads over the one or more disks; and
a controller coupled to the first and second actuators and the first and second heads, and operable to:
defining a first volume on a first portion of the one or more disks, reading from and writing to the first portion by the first read/write head;
defining a second volume on a second portion of the one or more disks, read from and written to the second portion by the second read/write head;
in response to writing data to the first volume, copying the data to the second drive volume; and
performing a background authentication operation on the second volume instead of the first volume.
Example 16. the system of example 15, wherein the first actuator and the second actuator move about separate pivots, and wherein at least a portion of the first roll and at least a portion of the second roll are located on a same surface of the one or more disks.
Example 17. the system of examples 15 or 16, wherein the second head is capable of being read from the first volume in parallel with the first head to increase throughput of host data requests.
Example 18. the system of example 15, wherein the second volume is a mirror image of the first volume.
The system of example 19, the system of example 15, wherein copying the data to the second volume comprises: compressing the data stored on the second volume.
Example 20. the system of example 15, wherein a host computer is given full access to the first volume and limited access to the second volume.
- 上一篇:一种医用注射器针头装配设备
- 下一篇:磁盘装置用的热致动器控制值的设定方法