Managing data input/output operations

• US 9,201,677 B2
• Filed: 07/27/2011
• Issued: 12/01/2015
• Est. Priority Date: 05/23/2011
• Status: Active Grant

• Current Assignee: SanDisk Technologies LLC (Western Digital Corporation)
• Original Assignee: Intelligent Intellectual Property Holdings LLC
• Inventors: Joshi, Vikram, Luan, Yang, Apte, Manish R., Vidwans, Hrishikesh A., Brown, Michael F.
• Primary Examiner(s): Bragdon, Reginald
• Assistant Examiner(s): Namazi, Mehdi
• Application Number:
  US13/192,365
• Publication Number:
  US 20120304171A1
• Time in Patent Office: 1,588 Days
• Field of Search: 711/6, 711/170, 711/171, 711/172, 711/173, 711/103, 711/129, 711/130, 711/147, 711/152, 711/153, 718/104
• US Class Current: 1/1
• CPC Class Codes: G06F 12/084 with a shared cache
• CPC Class Codes: G06F 12/0871 Allocation or management of cache space
• CPC Class Codes: G06F 2009/45579 I/O management, e.g. providing access to device drivers or storage
• CPC Class Codes: G06F 2009/45583 Memory management, e.g. access or allocation
• CPC Class Codes: G06F 2212/152 Virtualized environment, e.g. logically partitioned system
• CPC Class Codes: G06F 2212/601 Reconfiguration of cache memory
• CPC Class Codes: G06F 2212/6042 Allocation of cache space to multiple users or processors
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects

Abstract

Systems and methods for managing data input/output operations are described that include virtual machines operating with a shared storage within a host. In such a system, a computer-implemented method is provided for dynamically provisioning cache storage while operating system applications continue to operate, including stalling the virtual machine'"'"'s local cache storage operations, changing the provision of cache storage size; and resuming the operations of the virtual machine.

Claims

1. A method, comprising:

provisioning respective amounts of cache resources to virtual machines operating on a host computing device, such that a first one of the virtual machines is provisioned a first amount of cache resources and a second one of the virtual machines is provisioned a second, different amount of cache resources;

modifying the cache resources provisioned to the first virtual machine by assigning a set of cache storage locations comprising data of another virtual machine to the first virtual machine; and

servicing storage requests of the virtual machines by use of cache storage resources of the host computing device in accordance with the cache resources provisioned to the respective virtual machines, wherein servicing the storage requests of the virtual machines comprises;

blocking the first virtual machine from reading data stored on the set of cache storage locations;

allowing read access to a cache storage location of the set of cache storage locations in response the first virtual machine writing data to the cache storage location.

2. The method of claim 1, further comprising modifying the amount of cache resources provisioned to the first virtual machine in response to storage requests of the virtual machines.

3. The method of claim 1, further comprising:

monitoring input/output requests issued by the virtual machines; and

modifying the amount of cache resources provisioned to the respective virtual machines in response to determining one or more input/output requirements of the virtual machines based on the monitoring.

4. The method of claim 1, further comprising modifying the amount of cache resources provisioned to the first virtual machine based on a request from another one of the virtual machines.

5. The method of claim 1, wherein the cache resources provisioned to the virtual machines correspond to cache storage resources of a cache storage device, the method further comprising mapping cache addresses of respective virtual machines to physical storage locations of the cache storage device.

6. The method of claim 1, further comprising modifying the amount of cache resources provisioned to the first virtual machine by allocating a cache storage location of the first virtual machine to the second virtual machine, the method further comprising preventing the second virtual machine from accessing data of the first virtual machine stored on the cache storage location.

7. The method of claim 1, further comprising:

maintaining a bitmap representing write operations by the first virtual machine within the set of cache storage locations; and

identifying cache storage locations comprising data of the first virtual machine by use of the bitmap.

8. The method of claim 1, wherein provisioning the respective amounts of cache resources to the virtual machines comprises provisioning respective shares of cache storage capacity to the virtual machines within a cache.

9. The method of claim 1, wherein provisioning the respective amounts of cache resources to the virtual machines comprises provisioning respective shares of input/output capacity to the virtual machines.

10. An apparatus, comprising:

a virtual storage driver embodied as computer-readable instructions stored on a non-transitory storage medium of a computing device, the virtual storage driver configured to provide caching services to a plurality of virtual machines by use of a non-volatile storage device;

a cache provisioner module configured to allocate storage capacity of the non-volatile storage device to respective virtual machines for storage of cache data of the respective virtual machines; and

a cache map module configured to map virtual cache addresses of the virtual machines to storage addresses on the non-volatile storage device, and to control access to data of the virtual machines,

wherein the cache provisioner module is configured to assign a cache storage location to a particular virtual machine, and to block the particular virtual machine from reading data stored on the cache storage location until an operation to write data of the particular virtual machine to the cache storage location.

11. The apparatus of claim 10, wherein the cache provisioner module is configured to allocate a portion of the non-volatile storage device comprising cache data of a first one of the virtual machines to a second one of the virtual machines, and wherein the cache mapping module is configured to reject access, by the second virtual machine, to the cache data of the first virtual machine stored in the portion.

12. The apparatus of claim 11, wherein the cache mapping module is configured to provide access to the portion in response to storing cache data of the second virtual machine at least once within the portion.

13. The apparatus of claim 10, wherein the cache provisioner module is configured to dynamically modify the storage capacity of the non-volatile storage device allocated to the respective virtual machines, the apparatus further comprising a virtual disk configured to represent the dynamically allocated storage capacity as fixed-size storage devices within the respective virtual machines.

14. The apparatus of claim 10, wherein the virtual storage driver comprises a virtual logical unit number (VLUN) driver configured for operation within a virtualization kernel.

15. The apparatus of claim 10, wherein the cache provisioner module is configured to modify the storage capacity of the non-volatile storage device allocated to a first one of the virtual machines in response to storage operations of the virtual machines performed on the non-volatile storage device.

16. The apparatus of claim 10, wherein the cache provisioner module is configured to modify the storage capacity of the non-volatile storage device allocated to a first one of the virtual machines in response to an allocation request from one or more of the first virtual machine and another one of the virtual machines.

17. The apparatus of claim 10, wherein the cache provisioner module is configured to allocate input/output capacity between the respective virtual machines, and to manage an input/output rate of a virtual machine in response to determining that the virtual machine is exceeding an input/output capacity allocated to the virtual machine.

18. A non-transitory computer readable storage medium having program instructions stored thereon, wherein the program instructions are executable by a computing system to cause the computing system to perform operations, comprising:

assigning chunks of shared cache storage to respective virtual machines;

modifying a number of chunks allocated to the respective virtual machines in response to cache storage requirements of the virtual machines, wherein modifying the number of chunks allocated to the respective virtual machines comprises,

assigning a chunk of the shared virtual machine cache previously assigned to another virtual machine to a first virtual machine, and

restricting access, by the first virtual machine, to portions of the chunk corresponding to write operations of the first virtual machine; and

representing variably sized cache storage resources corresponding to the chunks of the shared cache storage allocated to the virtual machines as fixed-size storage devices within the respective virtual machines.

19. The non-transitory computer readable storage medium of claim 18, the operations further comprising mapping a contiguous set of virtual cache addresses of a virtual machine to a disjoint set of cache chunks of the shared cache storage.

20. The non-transitory computer readable storage medium of claim 18, the operations further comprising:

allocating input/output operations (IOPs) capacity to the virtual machines; and

servicing cache storage requests of the virtual machines in accordance with the IOPs capacity allocated to the virtual machines.

21. A method, comprising:

allocating input/output operations (IOPs) capacity to virtual machines operating within a hypervisor, such that a first one of the virtual machines is allocated a first IOPs capacity and a second one of the virtual machines is allocated a second, different IOPs capacity;

allocating cache storage capacity to the virtual machines from cache storage resources of the host computing device, including allocating a cache page to a first virtual machine that comprises data of a second virtual machine;

servicing storage requests of the virtual machines by use of the cache storage capacity allocated to the virtual machines in the cache storage resources of the host computing device in accordance with the I0Ps capacity allocated to the virtual machines, wherein servicing storage requests comprises restricting read access to the cache page until data is written to the cache page by the first virtual machine.