Method, apparatus, and computer readable medium for I/O control
Techniques providing I/O control involve: in response to receiving an I/O request, detecting a first set bits for a stripe in a RAID. The RAID is built on disk slices divided from disks. The stripes include extents. Each of the first set bits indicates whether a disk slice where a corresponding extent in the stripe is located is in a failure state. The techniques further involve determining, from the stripe and based on the first set bits, a first set of extents in the failure state and a second set of extents out of the failure state. The techniques further involve executing the I/O request on the second set of extents without executing the I/O request on the first set of extents. Such techniques can simplify storage bits in I/O control, support the degraded stripe write request for the RAID and enhance performance executing the I/O control.
1. A method for I/O control, comprising the steps, performed for a write request, of:
in response to receiving the write request, detecting a first set of information bits for a stripe in a Redundant Array of Independent Disks (RAID) related to the write request, the RAID being built on a plurality of disk slices divided from a plurality of disks, the stripe comprising a plurality of extents, and each of the first set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located is in a failure state, the failure state being a condition in which a disk slice has failed but has not yet been replaced;
determining, from the stripe and based on the first set of information bits, a first set of extents in the failure state and a second set of extents out of the failure state;
based on the determining, and without allocating a new disk slice to replace a disk slice indicated to be in the failure state, executing the write request as a first degraded write request on the second set of extents without executing the first degraded write request on the first set of extents;
detecting whether data of the first degraded write request is successfully written into the second set of extents; and
in response to determining that the data of the first degraded write request fails to be written into one of the second set of extents:
(1) updating the first set of information bits such that an information bit, which corresponds to a failed disk slice where the one extent in the failure state is located, is updated to indicate that the one extent is included in the first set of extents and not in the second set of extents; and
(2) re-executing the write request as a second degraded write request based on the updated first set of information bits, the second degraded write request being executed on the second set of extents without executing the degraded write request on the first set of extents.
2. The method of claim 1 , wherein the write request is followed by a read request executed as a degraded read request on the second set of extents by:
detecting a second set of information bits for the stripe, each of the second set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located includes valid data;
determining, from the second set of extents and based on the second set of information bits, a third set of extents including the valid data and a fourth set of extents without any valid data; and
executing the read request as a degraded read request on the third set of extents without executing the read request on the fourth set of extents.
3. The method of claim 2 , wherein:
the second set of information bits are determined based on whether data associated with the write request preceding the read request is successfully written into a disk slice.
4. The method of claim 1 , wherein the write request is followed by a read request executed as a degraded read request on the second set of extents by:
detecting a third set of information bits for the stripe, each of the third set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located is in a new disk rebuild state;
determining, from the second set of extents and based on the third set of information bits, a fifth set of extents in the new disk rebuild state and a sixth set of extents out of the new disk rebuild state; and
executing the read request as a degraded read request on the sixth set of extents without executing the read request on the fifth set of extents.
5. The method of claim 2 , wherein the second set of information bits are stored in mapper metadata after completion of the write request, and the stored second set of information bits are used in the execution of the read request so as not to perform the read request on disk slices without any valid data.
6. An apparatus for I/O control, comprising:
a processor; and
a memory coupled to the processor and having instructions stored thereon, the instructions, when executed by the processor, causing the apparatus to perform acts for a write request comprising:
in response to receiving a-the write request, detecting a first set of information bits for a stripe in a Redundant Array of Independent Disks (RAID) related to the write request, the RAID being built on a plurality of disk slices divided from a plurality of disks, the stripe comprising a plurality of extents, and each of the first set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located is in a failure state, the failure state being a condition in which a disk slice has failed but has not yet been replaced;
determining, from the stripe and based on the first set of information bits, a first set of extents in the failure state and a second set of extents out of the failure state;
based on the determining, and without allocating a new disk slice to replace a disk slice indicated to be in the failure state, executing the write request as a first degraded write request on the second set of extents without executing the first degraded write request on the first set of extents;
detecting whether data of the first degraded write request is successfully written into the second set of extents; and
in response to determining that the data of the first degraded write request fails to be written into one of the second set of extents:
(1) updating the first set of information bits such that an information bit, which corresponds to a failed disk slice where the one extent in the failure state is located, is updated to indicate that the one extent is included in the first set of extents and not in the second set of extents; and
(2) re-executing the write request as a second degraded write request based on the updated first set of information bits, the second degraded write request being executed on the second set of extents without executing the degraded write request on the first set of extents.
7. The apparatus of claim 6 , wherein the write request is followed by a read request executed as a degraded read request by:
detecting a second set of information bits for the stripe, each of the second set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located includes valid data;
determining, from the second set of extents and based on the second set of information bits, a third set of extents including the valid data and a fourth set of extents without any valid data; and
executing the read request as a degraded read request on the third set of extents without executing the read request on the fourth set of extents.
8. The apparatus of claim 7 , wherein:
the second set of information bits are determined based on whether data associated with the write request preceding the read request is successfully written into a disk slice.
9. The apparatus of claim 6 , wherein the write request is followed by a read request executed as a degraded read request on the second set of extents by:
detecting a third set of information bits for the stripe, each of the third set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located is in a new disk rebuild state;
determining, from the second set of extents and based on the third set of information bits, a fifth set of extents in the new disk rebuild state and a sixth set of extents out of the new disk rebuild state; and
executing the read request as a degraded read request on the sixth set of extents without executing the read request on the fifth set of extents.
10. A non-transitory computer-readable medium having computer-readable instructions stored thereon, the computer-readable instructions, when executed, causing a device to perform acts for a write request including:
in response to receiving the write request, detecting a first set of information bits for a stripe in a Redundant Array of Independent Disks (RAID) related to the write request, the RAID being built on a plurality of disk slices divided from a plurality of disks, the stripe comprising a plurality of extents, and each of the first set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located is in a failure state, the failure state being a condition in which a disk slice has failed but has not yet been replaced;
determining from the stripe and based on the first set of information bits, a first set of extents in the failure state and a second set of extents out of the failure state;
based on the determining, and without allocating a new disk slice to replace a disk slice indicated to be in the failure state, executing the write request as a first degraded write request on the second set of extents without executing the first degraded write request on the first set of extents;
detecting whether data of the first degraded write request is successfully written into the second set of extents; and
in response to determining that the data of the first degraded write request fails to be written into one of the second set of extents:
(1) updating the first set of information bits such that an information bit, which corresponds to a failed disk slice where the one extent in the failure state is located, is updated to indicate that the one extent is included in the first set of extents and not in the second set of extents; and
(2) re-executing the write request as a second degraded write request based on the updated first set of information bits, the second degraded write request being executed on the second set of extents without executing the degraded write request on the first set of extents.
11. The non-transitory computer-readable medium of claim 10 , wherein the write request is followed by a read request, and the computer-readable instructions, when executed, cause the device to execute the read request as a degraded read request by:
detecting a second set of information bits for the stripe, each of the second set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located includes valid data;
determining, from the second set of extents and based on the second set of information bits, a third set of extents including the valid data and a fourth set of extents without any valid data; and
executing the read request as a degraded read request on the third set of extents without executing the read request on the fourth set of extents.
12. The non-transitory computer-readable medium of claim 11 , wherein:
the second set of information bits are determined based on whether data associated with the write request preceding the read request is successfully written into a disk slice.
13. The non-transitory computer-readable medium of claim 10 , wherein the write request is followed by a read request, and the computer-readable instructions, when executed, cause the device to execute the read request as a degraded read request by:
detecting a third set of information bits for the stripe, wherein each of the third set of information bits indicating whether a disk slice where a corresponding extent in the stripe is located is in a new disk rebuild state;
determining, from the second set of extents and based on the third set of information bits, a fifth set of extents in the new disk rebuild state and a sixth set of extents out of the new disk rebuild state; and
executing the read request as a degraded read request on the sixth set of extents without executing the read request on the fifth set of extents.