Method and system for dynamic write-back cache sizing in solid state memory storage

• US 10,289,550 B1
• Filed: 12/30/2016
• Issued: 05/14/2019
• Est. Priority Date: 12/30/2016
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Inventors: Van Gaasbeck, Richard H.
• Primary Examiner(s): Rones, Charles
• Assistant Examiner(s): Li, Hewy H
• Application Number:
  US15/394,944
• Time in Patent Office: 865 Days
• CPC Class Codes: G06F 12/0804 with main memory updating G06F12/0806 takes precedence
• CPC Class Codes: G06F 12/0868 Data transfer between cache memory and other subsystems, e.g. storage devices or host systems
• CPC Class Codes: G06F 12/0871 Allocation or management of cache space
• CPC Class Codes: G06F 2212/621 Coherency control relating to peripheral accessing, e.g. from DMA or I/O device

Abstract

A method for dynamic write-back cache sizing in solid state memory storage. The method includes receiving a request to write a data fragment to a memory location of a storage medium, identifying a resource required for the writing, and obtaining a state of the resource. The state of the resource is governed by a number of data fragments that are stored in the write-back cache which require the resource. The number of data fragments in the write-back cache which require the resource are limited to enable writing of all data fragments in the write-back cache to the storage medium, within a specified amount of time. The method further includes determining that the state of the resource allows the received data fragment to be stored in the write-back cache, and based on the determination: storing the data fragment in the write-back cache and acknowledging the write request.

Claims

1. A method for dynamic write-back cache sizing in solid state memory storage, the method comprising:

receiving a first request to write a first data fragment to a first memory location in solid state memory of a storage medium, wherein the first memory location is identified by a first write address of the solid state memory;

identifying, based on the first write address, a first resource required for writing the first data fragment to the first memory location;

obtaining a state of the first resource,

wherein the state of the first resource is governed by a number of data fragments that are currently stored in the write-back cache which require the first resource, and

wherein the number of data fragments that are currently stored in the write-back cache which require the first resource are limited to a number that enables a writing of all data fragments in the write-back cache to the storage medium, within a specified amount of time;

making a first determination that the state of the first resource allows the received first data fragment to be stored in the write-back cache, and based on the first determination;

storing the first data fragment in the write-back cache;

updating the state of the state of the first resource; and

acknowledging the first write request.

2. The method of claim 1, further comprising:

writing the first data fragment, stored in the write-back cache, to the storage medium, using the first resource; and

updating the state of the first resource.

3. The method of claim 1, further comprising:

receiving a second request to write a second data fragment to a second memory location on the storage medium, wherein the second memory location is identified by a second write address of the solid state memory;

identifying, based on the second write address, a second resource, required for writing the second data fragment to the second memory location;

obtaining a state of the second resource;

making a second determination that the state of the second resource does not allow the received data fragment to be stored in the write-back cache, and based on the second determination;

performing one selected from a group consisting of;

denying the storage of the second data fragment in the write-back cache and not acknowledging the second write request; and

delaying the storage of the second data fragment in the write-back cache and delaying acknowledging the second write request.

4. The method of claim 3, wherein, after the second write request is not acknowledged, the second data fragment is stored in a buffer of the solid state memory storage.

5. The method of claim 1, further comprising:

receiving a second request to write a second data fragment to a second memory location on the storage medium, wherein the second memory location is identified by a second write address of the solid state memory;

identifying the first and a second resource, required for writing the second data fragment to the second memory location, based on the second write address;

obtaining the state of the first resource;

making a second determination that the state of the first resource allows the received second data fragment to be stored in the write-back cache;

obtaining a state of the second resource;

making a third determination that the state of the second resource does not allow the second data fragment to be stored in the write-back cache; and

based on the second and the third determinations;

performing one selected from a group consisting of;

denying the storage of the second data fragment in the write-back cache and not acknowledging the second write request; and

delaying the storage of the second data fragment in the write-back cache and delaying acknowledging the second write request.

6. The method of claim 1,

wherein the state of the first resource is tracked using a counter;

wherein the counter is incremented when a data fragment whose writing to the storage medium requires the first resource is received by the write-back cache; and

wherein the counter is decremented when a data fragment, stored in the write-back cache, is written to the medium using the first resource.

7. The method of claim 1,

wherein the specified amount of time is governed by a time interval during which a backup power source powers the solid state memory storage after a loss of a main power source.

8. A storage appliance comprising:

a storage module comprising a storage module controller and a storage medium wherein the storage module controller is configured to;

receive a first request to write a first data fragment to a first memory location in solid state memory of the storage medium, wherein the first memory location is identified by a first write address of the solid state memory;

identify, based on the first write address, a first resource required for writing the first data fragment to the first memory location;

obtain a state of the first resource,

wherein the state of the first resource is governed by a number of data fragments that are currently stored in the write-back cache which require the first resource, and

wherein the number of data fragments that are currently stored in the write-back cache which require the first resource are limited to a number that enables a writing of all data fragments in the write-back cache to the storage medium, within a specified amount of time;

make a first determination that the state of the first resource allows the received first data fragment to be stored in the write-back cache, and based on the first determination;

store the first data fragment in the write-back cache;

update the state of the state of the first resource; and

acknowledge the first write request.

9. The storage appliance of claim 8, wherein the storage module controller is further configured to:

write the first data fragment, stored in the write-back cache, to the storage medium, using the first resource; and

update the state of the first resource.

10. The storage appliance of claim 8, wherein the storage module controller is further configured to:

receive a second request to write a second data fragment to a second memory location on the storage medium, wherein the second memory location is identified by a second write address of the solid state memory;

identify, based on the second write address, a second resource, required for writing the second data fragment to the second memory location;

obtain a state of the second resource;

make a second determination that the state of the second resource does not allow the received data fragment to be stored in the write-back cache, and based on the second determination;

perform one selected from the group consisting of;

denying the storage of the second data fragment in the write-back cache and not acknowledging the second write request; and

delaying the storage of the second data fragment in the write-back cache and delaying acknowledging the second write request.

11. The storage appliance of claim 10, wherein, after the second write request is not acknowledged, the second data fragment is stored in a buffer of the storage appliance.

12. The storage appliance of claim 8, wherein the storage module controller is further configured to:

receive a second request to write a second data fragment to a second memory location on the storage medium, wherein the second memory location is identified by a second write address of the solid state memory;

identify the first and a second resource, required for writing the second data fragment to the second memory location, based on the second write address;

obtain the state of the first resource;

make a second determination that the state of the first resource allows the received second data fragment to be stored in the write-back cache;

obtain a state of the second resource;

make a third determination that the state of the second resource does not allow the second data fragment to be stored in the write-back cache; and

based on the second and the third determinations;

perform one selected from a group consisting of;

denying the storage of the second data fragment in the write-back cache and not acknowledging the second write request; and

delaying the storage of the second data fragment in the write-back cache and delaying acknowledging the second write request.

13. The storage appliance of claim 8,

wherein the state of the first resource is tracked using a counter;

wherein the counter is incremented when a data fragment whose writing to the storage medium requires the first resource is received by the write-back cache; and

wherein the counter is decremented when a data fragment, stored in the write-back cache, is written to the medium using the first resource.

14. The storage appliance of claim 8,

wherein the specified amount of time is governed by a time interval during which a backup power source powers the storage module after a loss of a main power source.

15. A non-transitory computer readable medium (CRM) comprising instructions that enable a storage appliance to:

receive a first request to write a first data fragment to a first memory location in solid state memory of a storage medium, wherein the first memory location is identified by a first write address of the solid state memory;

identify, based on the first write address, a first resource required for writing the first data fragment to the first memory location;

obtain a state of the first resource,

wherein the state of the first resource is governed by a number of data fragments that are currently stored in the write-back cache which require the first resource, and

wherein the number of data fragments that are currently stored in the write-back cache which require the first resource are limited to a number that enables a writing of all data fragments in the write-back cache to the storage medium, within a specified amount of time;

make a first determination that the state of the first resource allows the received first data fragment to be stored in the write-back cache, and based on the first determination;

store the first data fragment in the write-back cache;

update the state of the state of the first resource; and

acknowledge the first write request.

16. The non-transitory CRM of claim 15 further comprising instructions that enable the storage appliance to:

write the first data fragment, stored in the write-back cache, to the storage medium, using the first resource; and

update the state of the first resource.

17. The non-transitory CRM of claim 15 further comprising instructions that enable the storage appliance to:

receive a second request to write a second data fragment to a second memory location on the storage medium, wherein the second memory location is identified by a second write address of the solid state memory;

identify, based on the second write address, a second resource, required for writing the second data fragment to the second memory location;

obtain a state of the second resource;

make a second determination that the state of the second resource does not allow the received data fragment to be stored in the write-back cache, and based on the second determination;

performing one selected from a group consisting of;

denying the storage of the second data fragment in the write-back cache and not acknowledging the second write request; and

delaying the storage of the second data fragment in the write-back cache and delaying acknowledging the second write request.

18. The non-transitory CRM of claim 17 wherein, after the second write request is not acknowledged, the second data fragment is stored in a buffer of the solid state memory storage.

19. The non-transitory CRM of claim 15 further comprising instructions that enable the storage appliance to:

receive a second request to write a second data fragment to a second memory location on the storage medium, wherein the second memory location is identified by a second write address of the solid state memory;

identify the first and a second resource, required for writing the second data fragment to the second memory location, based on the second write address;

obtain the state of the first resource;

make a second determination that the state of the first resource allows the received second data fragment to be stored in the write-back cache;

obtain a state of the second resource;

make a third determination that the state of the second resource does not allow the second data fragment to be stored in the write-back cache; and

based on the second and the third determinations;

performing one selected from a group consisting of;

denying the storage of the second data fragment in the write-back cache and not acknowledging the second write request; and

delaying the storage of the second data fragment in the write-back cache and delaying acknowledging the second write request.

20. The non-transitory CRM of claim 15,

wherein the state of the first resource is tracked using a counter;

wherein the counter is incremented when a data fragment whose writing to the storage medium requires the first resource is received by the write-back cache; and

wherein the counter is decremented when a data fragment, stored in the write-back cache, is written to the medium using the first resource.