Granted Patent B2

US 12,367,178 · App. 17/853,629 · Granted Jul 22, 2025

Inventors: Prabir Paul (San Jose, CA); Lakshmi Narasimhan Sundararajan (Tamil Nadu, IN); Nikhil Subhash Bhupale (Maharashtra, IN); Vinod Jayaraman (San Francisco, CA); Goutham Rao (Los Altos Hills, CA)

Assignee: Pure Storage, Inc.

G06F16/13G06F16/128G06F16/188G06F21/53G06F21/6218

Abstract

Claims (51)

1. A method of providing distributed persistent storage for containerized applications running in a container system, the method comprising:

virtualizing, by a container storage system and using a device mapper, data blocks of a storage device to form a virtual block device;

generating, by the container storage system and based on the virtual block device, a storage pool backing datastore in device namespace; and

creating, by the container storage system within the storage pool backing datastore, thin-provisioned volumes for use by the containerized applications running in the container system;

wherein the container storage system is implemented as a containerized application running in the container system.

2. The method of claim 1 , wherein the device mapper comprises DM-thin.

3. The method of claim 1 , wherein the thin-provisioned volumes are in the device namespace.

4. The method of claim 1 , further comprising snapshotting the thin-provisioned volumes.

5. The method of claim 1 , further comprising replicating the thin-provisioned volumes.

6. The method of claim 1 , wherein virtualizing the data blocks of the storage device to form the virtual block device comprises:

creating a RAID device from the data blocks of the storage device; and

creating a logical data volume and a logical metadata volume on the RAID device, the logical data volume and the logical metadata volume together forming the storage pool backing datastore.

7. The method of claim 1 , wherein virtualizing the data blocks of the storage device to form the virtual block device comprises:

creating a RAID device from the data blocks of the storage device to form a physical volume;

using the physical volume to create a volume group;

using the volume group to create a logical data volume and a logical metadata volume; and

combining the logical data volume and the logical metadata volume to form the storage pool backing datastore.

8. The method of claim 1 , further comprising:

receiving a request to modify a designated file included in a containerized application that is loaded into a memory module on a computing device and being executed by a processor at the computing device, the containerized application including instructions arranged in a plurality of layers, the layers being ordered from lowest to highest, each layer including a respective plurality of files, the designated file included in more than one of the layers;

identifying a snapshot corresponding to the designated file, the snapshot including one or more snapshot data blocks, each of the snapshot data blocks including data that is additional to or different from any data associated with the designated file in a lower level layer; and

modifying the snapshot instead of copying the designated file.

9. A system comprising:

one or more memories storing computer-executable instructions; and

one or more processors to execute the computer-executable instructions to:

virtualize, using a device mapper, data blocks of a storage device to form a virtual block device;

generate, based on the virtual block device, a storage pool backing datastore; and

create, within the storage pool backing datastore, thin-provisioned volumes for use by containerized applications running in a container system;

wherein the container storage system is implemented as a privileged containerized application running in the container system.

10. The system of claim 9 , wherein the device mapper comprises DM-thin.

11. The system of claim 9 , wherein the storage pool backing datastore is in device namespace.

12. The system of claim 9 , wherein the thin-provisioned volumes are in device namespace.

13. The system of claim 9 , further comprising snapshotting the thin-provisioned volumes.

14. The system of claim 9 , further comprising replicating the thin-provisioned volumes.

15. The system of claim 9 , wherein virtualizing the data blocks of the storage device to form the virtual block device comprises:

creating a RAID device from the data blocks of the storage device; and

creating a logical data volume and a logical metadata volume on the RAID device, the logical data volume and the logical metadata volume together forming the storage pool backing datastore.

16. The system of claim 9 , wherein virtualizing the data blocks of the storage device to form the virtual block device comprises:

creating a RAID device from the data blocks of the storage device to form a physical volume;

using the physical volume to create a volume group;

using the volume group to create a logical data volume and a logical metadata volume; and

combining the logical data volume and the logical metadata volume to form the storage pool backing datastore.

17. A non-transitory, computer-readable medium storing computer instructions that, when executed, direct one or more processors of one or more computing devices to:

virtualize, using a device mapper, data blocks of a storage device to form a virtual block device;

generate, based on the virtual block device, a storage pool backing datastore in device namespace; and

create, within the storage pool backing datastore, thin-provisioned volumes for use by containerized applications running in a container system;

wherein the container storage system is implemented as a containerized application running in the container system.

18. The non-transitory, computer-readable medium of claim 17 , wherein the device mapper comprises DM-thin.

19. The non-transitory, computer-readable medium of claim 17 , wherein the thin-provisioned volumes are in the device namespace.

20. The non-transitory, computer-readable medium of claim 17 , wherein virtualizing the data blocks of the storage device to form the virtual block device comprises:

creating a RAID device from the data blocks of the storage device; and

creating a logical data volume and a logical metadata volume on the RAID device, the logical data volume and the logical metadata volume together forming the storage pool backing datastore.

Assignments (2)

Continuity (5)

References Cited (109)