Primary data storage system with quality of service

• US 10,282,101 B2
• Filed: 11/02/2015
• Issued: 05/07/2019
• Est. Priority Date: 11/07/2011
• Status: Expired due to Fees

• Current Assignee: Pivot3, Inc., NexGen Storage, Inc. (Western Digital Corporation)
• Original Assignee: NexGen Storage, Inc. (Western Digital Corporation)
• Inventors: Gallant, David A., Long, Kelly E., Ashmore, Paul A., Sobolewski, Sebastian P.
• Primary Examiner(s): Yi, David
• Assistant Examiner(s): Ahmed, Zubair
• Application Number:
  US14/930,614
• Publication Number:
  US 20160117103A1
• Time in Patent Office: 1,282 Days
• Field of Search: 711114, 711112
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0614 Improving the reliability of storage systems
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0649 Lifecycle management
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• CPC Class Codes: G06F 3/0685 Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: G06F 7/24 Sorting, i.e. extracting data from one or more carriers, rearranging the data in numerical or other ordered sequence, and rerecording the sorted data on the original carrier or on a different carrier or set of carriers sorting methods in generalG06F7/36 takes precedence

Abstract

The invention is directed to a primary data storage system for use in a computer network in which a network allows user computers to transfer data to/from the primary data storage system. In one embodiment, the primary data storage system allows an administrator of the computer network to define two or more volumes on the primary data storage system and define quality of service goals for each volume. The primary data storage system operates so as allocate resources within the primary data storage system to the volumes based upon criticality and performance goals specified for each of the volumes.

Claims

1. A data storage system having a quality of service capability, the system comprising:

an input/output port configured to receive a block command packet that embodies one of a read block command and a write block command and transmitting a block result packet in reply to a block command packet;

a data store system having at least first and second data stores each configured to receive and store data in response to a write block command and retrieve and provide data in response to a read block-command;

wherein the first data store has first data storage characteristics;

wherein the second data store has second data storage characteristics;

wherein the data store system has a data store system quality of service goal;

wherein the data store system is adaptable to accommodating at least a first volume with a first volume quality of service goal and a second volume with a second volume quality of service goal;

a statistics database configured to receive, store, and provide data for use in making decisions related to the pursuit of the first volume quality of service goal, the second volume of service goal, and the data store system quality of service goal; and

a sorting processor configured to sort an input string comprised of multiple read/write block commands, wherein the sorting processor is configured to order the multiple read/write block commands in an output string based on;

(a) the first and second data storage characteristics of the first and second data stores, (b) the data store system quality of service goal, (c) the first volume quality of service goal, (d) the second volume quality of service goal and (e) statistical data provided by the statistics database, wherein the sorting processor is also configured, in connection with the sorting of a read/write block command, to determine which one of the first and second data stores should receive the read/write block command such that processing of the command via the selected one of the first and second data stores is unlikely to violate a time constraint.

2. A data storage system, as claimed in claim 1, wherein the sorting processor, in sorting a read/write block command, is capable of acquiring a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) statistical data.

3. A data storage system, as claimed in claim 1, wherein the sorting processor, in sorting a read/write block command, is capable of acquiring a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) statistical data associated with the first volume.

4. A data storage system, as claimed in claim 1, wherein the sorting processor, in sorting a read/write block command, is capable of acquiring:

a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) statistical data associated with whichever one of the first and second volumes the read/write block command is associated; and

a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) statistical data associated with the other of the first and second volumes in deciding the sorting of the read/write block command.

5. A data storage system, as claimed in claim 1, wherein the sorting processor, in sorting read/write block commands into the output string, is capable of accessing a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) statistical data relating to the first and second data stores.

6. A data storage system, as claimed in claim 1, wherein the sorting processor is capable of determining a priority of a read/write block command relative to other read/write block commands using a weighted sum of weighted factors, the factors comprising:

(a) the criticality of the volume with which the read/write command is associated;

(b) the relationship of the read/write block command to the quality of service goal for the volume with which the read/write command is associated;

(c) the relationship of the read/write block command to the quality of service goal for any other volume having the same criticality as the volume with which the read/write command is associated;

(d) the relationship of the read/write block command to the first and second data stores; and

(e) the relationship of the read/write block command to the quality of service goal for any other volume having a different criticality than the volume with which the read/write command is associated.

7. A data storage system, as claimed in claim 6, wherein at least one factor has a value that is a current statistic value.

8. A data storage system, as claimed in claim 6, wherein at least one factor has a value that is a historic statistic value.

9. A data storage system, as claimed in claim 6, wherein each factor is weighted by a coefficient whose value relates to:

(a) the criticality and (b) reducing the difference between quality of service goals for the first and second volumes and the actual service obtained.

10. A data storage system, as claimed in claim 6, wherein each factor is weighted by a coefficient that changes over time.

11. A data storage system, as claimed in claim 1, wherein the statistics database is capable of providing the sorting processor with current statistical data that relates to a first time frame and historical statistical data that relates to a second time frame that is greater than the first time frame.

12. A data storage system, as claimed in claim 11, wherein the current statistical data and the historical statistical data relate to the first volume.

13. A data storage system, as claimed in claim 11, wherein the current statistical data and the historical statistical data relate to a first criticality associated with the first volume.

21. A data storage system, as claimed in claim 1, wherein the data store system quality of service goal includes an input-output operations per second (IOPS) goal.

14. A data storage system having a quality of service capability, the system comprising:

an input/output port configured to receive a block command packet that embodies one of a read block command and a write block command and transmitting a block result packet in reply to a block command packet;

a data store system having at least first and second data stores each configured to receive and store data in response to a write block command and retrieve and provide data in response to a read block-command;

wherein the first data store has first data storage characteristics;

wherein the second data store has second data storage characteristics;

wherein the data store system has a data store system quality of service goal;

the data store system is adaptable to accommodating at least a first volume with a volume quality of service goal; and

a sorting processor configured to sort an input string comprised of multiple read/write block commands, wherein the sorting processor is configured to order the multiple read/write block commands in an output string based on the;

(a) first and second data storage characteristics of the first and second data stores (b) the data store system quality of service goal and (c) the volume quality of service goal, wherein the sorting processor is also configured, in connection with the sorting of a read/write block command, to determine which one of the first and second data stores should receive the read/write block command such that processing of the command via the selected one of the first and second data stores is unlikely to violate a time constraint.

15. A data storage system, as claimed in claim 14, wherein the sorting processor, in sorting a read/write block command, is capable of acquiring a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) characteristics.

16. A data storage system, as claimed in claim 14, wherein the sorting processor, in sorting a read/write block command, is capable of acquiring a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) related to the volume quality of service goal associated with the first volume.

17. A data storage system, as claimed in claim 14, wherein the sorting processor, in sorting read/write block commands into the output string, is capable of accessing a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) characteristics relating to the first and second data stores.

18. A data storage system, as claimed in claim 14, wherein the sorting processor capable of determining a priority of a read/write block command relative to other read/write block commands using a weighted sum of weighted factors, the factors comprising:

(a) the criticality of the volume with which the read/write command is associated;

(b) the relationship of the read/write block command to the quality of service goal for the volume with which the read/write command is associated;

(c) the relationship of the read/write block command to the quality of service goal for any other volumes having the same criticality as the volume with which the read/write command is associated;

(d) the relationship of the read/write block command to the first and second data stores; and

(e) the relationship of the read/write block command to the quality of service goal for any other volumes having a different criticality than the volume with which the read/write command is associated.

19. A data storage system, as claimed in claim 18, wherein each factor is weighted by a coefficient that changes over time.

20. A data storage system, as claimed in claim 14, wherein the sorting processor, in sorting a read/write block command, is capable of acquiring a combination of a criticality, throughput, queue depth, latency, and input-output operations per second (IOPS) related to the data store system quality of service goal.

22. A data storage system, as claimed in claim 14, wherein the data store system quality of service goal includes an input-output operations per second (IOPS) goal.

23. A data storage system, as claimed in claim 14, further comprising a statistics database configured to receive, store, and provide data for use in making decisions related to the pursuit of the volume quality of service goal and the data store system quality of service goal.