# US 2015/0106563 A1 · Efficient support for drives with different sector alignments in a raid layout

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2015/0106563 A1
- Document ID: US20150106563A1
- Status: Published application
- Kind: A1
- Application number: 14/054,627
- Filed: 2013-10-15
- Published: 2015-04-16
- Earliest claimed: 2013-10-15
- Assignees: Dell Products L.P.
- Applicants: Dell Products, L.P.
- Inventors: Damon Hsu-Hung; Jeffrey L. Grummon
- Web page: https://patentgrep.com/patent/US20150106563A1

## CPC

- G06F 3/0614: {Improving the reliability of storage systems}
- G06F 3/0626: {Reducing size or complexity of storage systems}
- G06F 3/0634: {by changing the state or mode of one or more devices}
- G06F 3/064: {Management of blocks}
- G06F 3/0644: {Management of space entities, e.g. partitions, extents, pools}
- G06F 3/0689: {Disk arrays, e.g. RAID, JBOD}
- G06F 11/00: Error detection; Error correction; Monitoring (error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00)
- G06F 11/1096: {Parity calculation or recalculation after configuration or reconfiguration of the system}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

In one embodiment, a method includes receiving an input/output (I/O) request for data that starts or ends at a location other than a physical sector boundary of the device. The method further includes reading, starting at a first physical sector boundary before a beginning location specified in the I/O request and ending at a second physical sector boundary after an ending location specified in the request.

## Claims

### Claim 1

A method comprising:
responsive to a write request to a redundant array of independent drives (RAID) with a plurality of drives, determining a RAID operation to fulfill the write request, and further determining a plurality of data items required to complete the RAID operation, wherein each data item comprises a data range on one of the plurality of drives, wherein the data range specified in each data item may start or end at a location other than a physical sector boundary of the drive specified in the data item;
pre-reading at least one of the plurality of data items, starting at a first physical sector boundary at or before a beginning location specified in each corresponding data range and ending at a second physical sector boundary at or after an ending location specified in each corresponding data range;
performing at least one of a modify operation to generate new parity or mirror data based on data from the write request and data from the pre-reads, a modify operation to overlay data from the write request onto the data from the pre-reads, and a modify operation to overlay data from the parity calculation onto data from the pre-reads; and
writing any data item containing data from the write request starting at the first physical sector boundary and ending at the second physical sector boundary;
wherein the plurality of disk drives are enabled to have one or a plurality of physical sector sizes.

### Claim 2 (depends on 1)

The method of claim 1, wherein the determined RAID operation can be at least one of an additive stripe update, a subtractive stripe update, and a mirrored stripe update.

### Claim 3 (depends on 1)

The method of claim 1, wherein the determined RAID operation is a mirrored stripe update, pre-reading includes selecting one of the plurality of data items to read.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
providing one or more logical volumes having one or more logical sector sizes;
wherein the disks in the RAID are capable of having different physical sector sizes.

### Claim 5 (depends on 4)

The method of claim 4, wherein the particular physical block boundary scheme is at least one of 512 bytes and 4096 bytes, and the particular virtualized block boundary scheme is at least one of 512 bytes and 4096 bytes.

### Claim 6 (depends on 4)

The method of claim 4, wherein an internal block boundary scheme represents a common divisor of all of the physical sector sizes and all of the logical sector sizes.

### Claim 7 (depends on 6)

The method of claim 6, wherein the common divisor is the greatest common divisor.

### Claim 8 (depends on 4)

The method of claim 4, wherein a logical volume is hosted on one or more nodes, each node with a corresponding RAID, the physical sector sizes in a first of the nodes being able to differ from the physical sector sizes in a second node, wherein at least one of the nodes includes the RAID with the plurality of drives.

### Claim 9 (depends on 8)

The method of claim 8, further comprising:
moving data having a logical address that maps to a first physical address from a first disk in the RAID, the first disk having a first physical block boundary scheme or first virtualized block boundary scheme, to a second physical address on a second disk in the RAID having a second physical block boundary scheme or second virtualized block boundary scheme; and
mapping the logical address of data to the second physical address on the second disk.

### Claim 10

A system comprising:
a request analysis module configured to, responsive to a write request to a redundant array of independent drives (RAID) with a plurality of drives, determine a RAID operation to fulfill the write request, and further determine a plurality of data items required to complete the RAID operation, wherein each data item comprises a data range on one of the plurality of drives, wherein the data range specified in each data item may start or end at a location other than a physical sector boundary of the drive specified in the data item;
a drive reading module configured to pre-read at least one of the plurality of data items, starting at a first physical sector boundary at or before a beginning location specified in each corresponding data range and ending at a second physical sector boundary at or after an ending location specified in each corresponding data range;
a modification module configured to perform at least one of a modify operation to generate new parity or mirror data based on data from the write request and data from the pre-reads, a modify operation to overlay data from the write request onto the data from the pre-reads, and a modify operation to overlay data from the parity calculation onto data from the pre-reads; and
a disk writing module configured to write any data item containing data from the write request starting at the first physical sector boundary and ending at the second physical sector boundary;
wherein the plurality of disk drives are enabled to have one or a plurality of physical sector sizes.

### Claim 11 (depends on 10)

The method of claim 10, wherein the determined RAID operation can be at least one of an additive stripe update, a subtractive stripe update, and a mirrored stripe update.

### Claim 12 (depends on 10)

The method of claim 10, wherein the determined RAID operation is a mirrored stripe update, pre-reading includes selecting one of the plurality of data items to read.

### Claim 13 (depends on 10)

The method of claim 10, further comprising:
a presentation module configured to provide one or more logical volumes having one or more logical sector sizes;
wherein the disks in the RAID are capable of having different physical sector sizes.

### Claim 14 (depends on 13)

The method of claim 13, wherein the particular physical block boundary scheme is at least one of 512 bytes and 4096 bytes, and the particular virtualized block boundary scheme is at least one of 512 bytes and 4096 bytes.

### Claim 15 (depends on 13)

The method of claim 13, wherein an internal block boundary scheme represents a common divisor of all of the physical sector sizes and all of the logical sector sizes.

### Claim 16 (depends on 15)

The method of claim 15, wherein the common divisor is the greatest common divisor.

### Claim 17 (depends on 13)

The method of claim 13, wherein a logical volume is hosted on one or more nodes, each node with a corresponding RAID, the physical sector sizes in a first of the nodes being able to differ from the physical sector sizes in a second node, wherein at least one of the nodes includes the RAID with the plurality of drives.

### Claim 18 (depends on 8)

The method of claim 8, further comprising:
a data moving module configured to move data having a logical address that maps to a first physical address from a first disk in the RAID, the first disk having a first physical block boundary scheme or first virtualized block boundary scheme, to a second physical address on a second disk in the RAID having a second physical block boundary scheme or second virtualized block boundary scheme; and
a mapping module configured to map the logical address of data to the second physical address on the second disk.

### Claim 19

A non-transitory computer-readable medium configured to store instructions for supporting input/output (I/O) requests to drives with different physical block sizes, the instructions, when loaded and executed by a processor, causes the processor to:
responsive to a write request to a redundant array of independent drives (RAID) with a plurality of drives, determine a RAID operation to fulfill the write request, and further determine a plurality of data items required to complete the RAID operation, wherein each data item comprises a data range on one of the plurality of drives, wherein the data range specified in each data item may start or end at a location other than a physical sector boundary of the drive specified in the data item;
pre-read at least one of the plurality of data items, starting at a first physical sector boundary at or before a beginning location specified in each corresponding data range and ending at a second physical sector boundary at or after an ending location specified in each corresponding data range;
perform at least one of a modify operation to generate new parity or mirror data based on data from the write request and data from the pre-reads, a modify operation to overlay data from the write request onto the data from the pre-reads, and a modify operation to overlay data from the parity calculation onto data from the pre-reads; and
write any data item containing data from the write request starting at the first physical sector boundary and ending at the second physical sector boundary;
wherein the plurality of disk drives are enabled to have one or a plurality of physical sector sizes.

### Claim 20 (depends on 19)

The non-transitory computer-readable medium of claim 19, wherein the determined RAID operation can be at least one of an additive stripe update, a subtractive stripe update, and a mirrored stripe update.
