Automatic read and write acceleration of data accessed by virtual machines

Patent metadata

Field	Value
Publication number	US-9699263-B1
Application number	US-201313831677-A
Country	US
Kind code	B1
Filing date	Mar 15, 2013
Priority date	Aug 17, 2012
Publication date	Jul 4, 2017
Grant date	Jul 4, 2017

Abstract

The various implementations described herein include methods and systems for automatic management of data access acceleration in a computer system executing a plurality of clients. The method includes: receiving data access commands from two or more clients to access data in objects identified by the data access commands; and processing the data access commands to update access history information for portions of the objects identified by the data access commands. The method further includes: in accordance with the access history information, automatically identifying and marking for acceleration the portions of the objects identified by the data access commands that satisfy an access based data acceleration policy; and accelerating the object portions marked for acceleration, including accelerating data writes and data reads of the object portions to and from the persistent cache, where the persistent cache is shared by the two or more clients.

First claim

What is claimed is:

1. A method for accelerating data access, performed by a computer system having one or more processors, memory, a tiered data structure stored in the memory, the tiered data structure comprising a plurality of tiers, and a persistent cache for storing accelerated data, the method comprising:
receiving, at the computer system, data access commands from two or more clients to access data in objects identified by the data access commands;
processing, by the computer system, the data access commands to update access history information for respective object portions of the objects identified by the data access commands from the two or more clients, wherein the processing includes:
in accordance with the access history information, determining usage rates for each of the respective object portions identified by the data access commands, and
assigning each of the respective object portions identified by the data access commands to one of the plurality of tiers within the tiered data structure in accordance with a determination that a respective determined usage rate for a respective object portion is within a range of usage rates for a respective tier to which the respective object portion is assigned;

wherein each tier of the plurality of tiers within the tiered data structure stores information identifying object portions assigned to that tier;
automatically identifying and marking for acceleration at least some of the respective object portions identified by the data access commands from the two or more clients in accordance with respective tiers to which the at least some of the respective object portions have been assigned, wherein the automatically identifying and marking are performed collectively for the at least some of the respective object portions identified by the data access commands from the two or more clients; and
accelerating data access to the at least some of the respective object portions marked for acceleration, by writing the at least some respective object portions marked for acceleration to the persistent cache;
wherein the persistent cache is shared by the two or more clients.

2. The method of claim 1 , wherein the persistent cache comprises non-volatile solid state storage.

3. The method of claim 1 , wherein the two or more clients comprise virtual machines executed by the computer system.

4. The method of claim 1 , wherein the objects comprise one or more virtual disks.

5. The method of claim 1 , wherein the respective object portions comprise respective address blocks or sub-blocks of the objects.

6. The method of claim 5 , wherein the access history information includes at least one usage history value for a respective address block and at least one distinct respective usage history value for each sub-block of the respective address block.

7. The method of claim 1 , wherein the access history information includes at least one operations count for each respective object portion of a plurality of object portions identified by the data access commands.

8. The method of claim 1 , wherein updating the access history information comprises:
incrementing a usage history value for a respective object portion assigned to a respective tier for which recent usage criteria are satisfied; and
decrementing a usage history value for other portions of objects assigned to the respective tier for which recent usage criteria are not satisfied.

9. A computer system, comprising:
one or more processors;
a persistent cache for storing accelerated data;
memory storing a tiered data structure, the tiered data structure comprising a plurality of tiers, and one or more programs for execution by the one or more processors, wherein the one or more programs include instructions that when executed by the one or more processors cause the computer system to:
receive, at the computer system, data access commands from two or more clients to access data in objects identified by the data access commands;
process, by the computer system, the data access commands to update access history information for respective object portions of the objects identified by the data access commands from the two or more clients, wherein the instructions for processing the data access commands include instructions that when executed by the one or more processors cause the computer system to:
in accordance with the access history information, determine usage rates for each of the respective object portions identified by the data access commands, and
assign each of the respective object portions identified by the data access commands to one of the plurality of tiers within the tiered data structure in accordance with a determination that a respective determined usage rate for a respective object portion is within a range of usage rates for a respective tier to which the respective object portion is assigned;

wherein each tier of the plurality of tiers within the tiered data structure stores information identifying object portions assigned to that tier;
automatically identify and mark for acceleration at least some of the respective object portions identified by the data access commands from the two or more clients in accordance with respective tiers to which the at least some of the respective object portions have been assigned, wherein the automatically identifying and marking are performed collectively for the at least some of the respective object portions identified by the data access commands from the two or more clients; and
accelerate data access to the at least some of the respective object portions marked for acceleration, by writing the at least some respective object portions marked for acceleration to the persistent cache;
wherein the persistent cache is shared by the two or more clients.

10. The computer system of claim 9 , wherein the two or more clients comprise virtual machines executed by the computer system.

11. The computer system of claim 9 , wherein the respective object portions comprise respective address blocks or sub-blocks of the objects.

12. The computer system of claim 11 , wherein the access history information includes at least one usage history value for a respective address block and at least one distinct respective usage history value for each sub-block of the respective address block.

13. The computer system of claim 9 , wherein the access history information includes at least one operations count for each respective object portion of a plurality of object portions identified by the data access commands.

14. A non-transitory computer readable medium storing one or more programs that when executed by one or more processors of a computer system cause the computer system to:
receive, at the computer system, data access commands from two or more clients to access data in objects identified by the data access commands;
process, by the computer system, the data access commands to update access history information for respective object portions of the objects identified by the data access commands from the two or more clients, wherein causing the computer system to process the data access commands includes causing the computer system to:
in accordance with the access history information, determine usage rates for each of the respective object portions identified by the data access commands, and
assign each of the respective object portions identified by the data access commands to one of the plurality of tiers within the tiered data structure in accordance with a determination that a respective determined usage rate for a respective object portion is within a range of usage rates for a respec

Assignees

• Sandisk Entpr Ip Llc
• Sandisk Technologies Llc

Inventors

• Shats Serge
• Jelvis Alexei
• Ratty Muthukumar
• Vignanker Eugene

Classifications

• G06F12/0804

  with main memory updating (G06F12/0806 takes precedence) · CPC title

• H04L67/2852

  Electricity · mapped topic

• G06F12/08

  in hierarchically structured memory systems, e.g. virtual memory systems · CPC title

• H04L67/2842 Primary

  Electricity · mapped topic

• G06F17/30902

  Physics · mapped topic