Systems and methods for distributing cache space

• US 10,452,553 B1
• Filed: 10/16/2017
• Issued: 10/22/2019
• Est. Priority Date: 10/16/2017
• Status: Active Grant

• Current Assignee: Veritas Technologies, LLC (Whitehouse Group Ltd.)
• Original Assignee: Veritas Technologies, LLC (Whitehouse Group Ltd.)
• Inventors: Mahajan, Ketan, Pendharkar, Niranjan, Khatpe, Mahesh
• Primary Examiner(s): Bataille, Pierre Miche
• Application Number:
  US15/784,679
• Time in Patent Office: 736 Days
• Field of Search: 711171, 711119
• CPC Class Codes: G06F 12/0842 for multiprocessing or multitasking
• CPC Class Codes: G06F 12/0871 Allocation or management of cache space
• CPC Class Codes: G06F 12/0888 using selective caching, e.g. bypass
• CPC Class Codes: G06F 12/0893 Caches characterised by their organisation or structure
• CPC Class Codes: G06F 12/122 of the least frequently used [LFU] type, e.g. with individual count value
• CPC Class Codes: G06F 12/128 adapted to multidimensional cache systems, e.g. set-associative, multicache, multiset or multilevel
• CPC Class Codes: G06F 2212/6042 Allocation of cache space to multiple users or processors
• CPC Class Codes: G06F 2212/6046 Using a specific cache allocation policy other than replacement policy
• CPC Class Codes: G06F 2212/621 Coherency control relating to peripheral accessing, e.g. from DMA or I/O device
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• CPC Class Codes: G06F 9/4843 by program, e.g. task dispatcher, supervisor, operating system

Abstract

The disclosed computer-implemented method for distributing cache space may include (i) identifying workloads that make input/output requests to a storage system that comprises a cache that stores a copy of data recently written to the storage system, (ii) calculating a proportion of the cache that is occupied by data written to the cache by a workload, (iii) determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate, and (iv) limiting the volume of input/output requests from workload that will be accepted by the storage system in response to determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate. Various other methods, systems, and computer-readable media are also disclosed.

Claims

1. A computer-implemented method for distributing cache space, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:

identifying a plurality of workloads that make input/output requests to a storage system that comprises a cache that stores a copy of data recently written to the storage system by the plurality of workloads;

calculating a proportion of the cache that is occupied by data written to the cache by a workload within the plurality of workloads, wherein calculating the proportion of the cache that is occupied by the data written to the cache by the workload comprises examining data currently stored in the cache to determine which workload within the plurality of workloads wrote each portion of data currently stored in the cache and monitoring a rate at which the workload writes data to the cache;

determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate; and

limiting the volume of input/output requests from workload that will be accepted by the storage system in response to determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate.

2. The computer-implemented method of claim 1, wherein determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate comprises determining that the data written to the cache by the workload occupies more than a predetermined threshold for a proportion of cache space occupied by data written to the cache by a single workload.

3. The computer-implemented method of claim 1, wherein limiting the volume of input/output requests from workload that will be accepted by the storage system comprises reducing a number of credits granted to the workload, wherein credits granted to the workload regulate the volume of input/output requests that the storage system will accept from the workload within a predetermined time period.

4. The computer-implemented method of claim 3, wherein reducing the number of credits granted to the workload comprises granting the workload a number of credits based at least in part on the inverse of the proportion of cache space occupied by the data written to the cache by the workload.

5. The computer-implemented method of claim 3, wherein reducing the number of credits granted to the workload comprises:

granting a number of credits to the workload based on a service level agreement for the workload; and

withholding an additional number of credits from the workload in response to determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate.

6. The computer-implemented method of claim 1, wherein limiting the volume of input/output requests from workload that will be accepted by the storage system comprises limiting the rate at which the workload writes data to the cache to be no greater than a rate at which the cache flushes the data written by the workload from the cache.

7. The computer-implemented method of claim 1, further comprising preserving a quality of service for at least one additional workload within the plurality of workloads by limiting the volume of input/output requests from workload that will be accepted by the storage system.

8. The computer-implemented method of claim 1, wherein the cache comprises a write-back cache.

9. The computer-implemented method of claim 1, further comprising rejecting an input/output request from the workload in response to determining that the workload has reached a limit for input/output requests to the storage system.

10. A system for distributing cache space, the system comprising:

an identification module, stored in memory, that identifies a plurality of workloads that make input/output requests to a storage system that comprises a cache that stores a copy of data recently written to the storage system by the plurality of workloads;

a calculation module, stored in memory, that calculates a proportion of the cache that is occupied by data written to the cache by a workload within the plurality of workloads, wherein calculating the proportion of the cache that is occupied by the data written to the cache by the workload comprises examining data currently stored in the cache to determine which workload within the plurality of workloads wrote each portion of data currently stored in the cache and monitoring a rate at which the workload writes data to the cache;

a determination module, stored in memory, that determines that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate;

a limit module, stored in memory, that limits the volume of input/output requests from workload that will be accepted by the storage system in response to determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate; and

at least one physical processor configured to execute the identification module, the calculation module, the determination module, and the limit module.

11. The system of claim 10, wherein the determination module determines that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate by determining that the data written to the cache by the workload occupies more than a predetermined threshold for a proportion of cache space occupied by data written to the cache by a single workload.

12. The system of claim 10, wherein the limit module limits the volume of input/output requests from workload that will be accepted by the storage system by reducing a number of credits granted to the workload, wherein credits granted to the workload regulate the volume of input/output requests that the storage system will accept from the workload within a predetermined time period.

13. The system of claim 12, wherein the limit module reduces the number of credits granted to the workload by granting the workload a number of credits based at least in part on the inverse of the proportion of cache space occupied by the data written to the cache by the workload.

14. The system of claim 12, wherein the limit module reduces the number of credits granted to the workload by:

granting a number of credits to the workload based on a service level agreement for the workload; and

withholding an additional number of credits from the workload in response to determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate.

15. The system of claim 10, wherein the limit module limits the volume of input/output requests from workload that will be accepted by the storage system by limiting the rate at which the workload writes data to the cache to be no greater than a rate at which the cache flushes the data written by the workload from the cache.

16. A non-transitory computer-readable medium comprising one or more computer-readable instructions that, when executed by at least one processor of a computing device, cause the computing device to:

identify a plurality of workloads that make input/output requests to a storage system that comprises a cache that stores a copy of data recently written to the storage system by the plurality of workloads;

calculate a proportion of the cache that is occupied by data written to the cache by a workload within the plurality of workloads, wherein calculating the proportion of the cache that is occupied by the data written to the cache by the workload comprises examining data currently stored in the cache to determine which workload within the plurality of workloads wrote each portion of data currently stored in the cache and monitoring a rate at which the workload writes data to the cache;

determine that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate; and

limit the volume of input/output requests from workload that will be accepted by the storage system in response to determining that the proportion of the cache that is occupied by the data written to the cache by the workload is disproportionate.