Optimized addition and removal of compute resources in a distributed storage platform by implementing mapping changes in a shared storage subsystem

Patent metadata

Field	Value
Publication number	US-2023051684-A1
Application number	US-202117401561-A
Country	US
Kind code	A1
Filing date	Aug 13, 2021
Priority date	Aug 13, 2021
Publication date	Feb 16, 2023
Grant date	—

Abstract

Computer-implemented methods for optimized compute resource addition and removal in a distributed storage platform. In a case of a newly added compute resource being connected to a storage subsystem shared by compute resources in the distributed storage platform, the distributed storage platform formulates a redistribution plan to redistribute a subset of a global address space of the storage subsystem to a newly added logical volume in the storage subsystem. In a case of a removed compute resource being disconnected from the storage subsystem, the distributed storage platform formulates a redistribution plan to redistribute respective logical blocks in a logical volume for the removed compute resource to respective remaining logical volumes for respective remaining compute resources in the distributed storage platform. The distributed storage platform executes the redistribution plan to reassign data block ownerships on one or more physical memory devices in the storage subsystem.

First claim

1 . A computer-implemented method for optimized addition of a compute resource in a distributed storage platform, the method comprising:
in response to that a newly added compute resource is connected to a storage subsystem shared by compute resources in the distributed storage platform, formulating a redistribution plan to redistribute a subset of a global address space of the storage subsystem to a newly added logical volume in the storage subsystem, wherein the newly added logical volume is assigned to the newly added compute resource; executing the redistribution plan to reassign data block ownerships on one or more physical memory devices in the storage subsystem, by modifying a mapping of respective logical block addresses in the storage subsystem to respective data block addresses on the one or more physical memory devices; and wherein the redistribution plan is a list of pairs, wherein each pair includes a source logical block address and a target logical block address, wherein the source logical block address is an address of a logical block in an existing logical volume assigned to an existing compute resource and the target logical block address is an address of a logical block in the newly added logical volume.

2 . The computer-implemented method of claim 1 , further comprising:
detecting addition of the newly added compute resource in the distributed storage platform.

3 . The computer-implemented method of claim 2 , further comprising:
in response to the newly added compute resource is detected, creating an empty logical volume as the newly added logical volume in the storage subsystem; and assigning the empty logical volume to the newly added compute resource.

4 . The computer-implemented method of claim 1 , further comprising:
modifying a mapping of logical block addresses in the global address space of the storage subsystem to the compute resources including the newly added compute resources.

5 . (canceled)

6 . The computer-implemented method of claim 1 , wherein N/(M+1) blocks from existing logical volumes assigned to existing compute resources in the distributed storage platform are redistributed to the newly added logical volume, wherein N is a number of data blocks in the storage subsystem, wherein M is a number of the existing logical volumes.

7 . The computer-implemented method of claim 1 , wherein, in modifying the mapping, an entry containing a logical block address that is redistributed according to the redistribution plan is modified.

8 . A computer-implemented method for optimized removal of a compute resource in a distributed storage platform, the method comprising:
in response to that a removed compute resource is disconnected from a storage subsystem shared by compute resources in the distributed storage platform, formulating a redistribution plan of the storage subsystem to redistribute respective logical blocks in a logical volume for the removed compute resource to respective remaining logical volumes for respective remaining compute resources in the distributed storage platform; executing the redistribution plan to reassign data block ownerships on one or more physical memory devices in the storage subsystem, by modifying a mapping of respective logical block addresses in the storage subsystem to respective data block addresses on the one or more physical memory devices; wherein the redistribution plan is a list of pairs, wherein each pair includes a source logical block address and a target logical block address, wherein the source logical block address is an address of a logical block in the logical volume for the removed compute resource and the target logical block address is an address of a logical block in a logical volume for a remaining compute resource.

9 . The computer-implemented method of claim 8 , further comprising:
detecting removal of the removed compute resource in the distributed storage platform.

10 . The computer-implemented method of claim 8 , further comprising:
modifying a mapping of logical block addresses in a global address space of the storage subsystem to the respective remaining compute resources.

11 . (canceled)

12 . The computer-implemented method of claim 8 , wherein, in modifying the mapping, an entry containing a logical block address that is redistributed according to the redistribution plan is modified.

13 . A computer-implemented method for optimized compute resource addition and removal in a distributed storage platform, the method comprising:
in response to that a newly added compute resource is connected to a storage subsystem shared by compute resources in the distributed storage platform, formulating a first redistribution plan of the storage subsystem to redistribute a subset of a global address space of the storage subsystem to a newly added logical volume in the storage subsystem, wherein the newly added logical volume is assigned to the newly added compute resource; in response to that a removed compute resource is disconnected from the storage subsystem shared by the compute resources in the distributed storage platform, formulating a second redistribution plan of the storage subsystem to redistribute respective logical blocks in a logical volume for the removed compute resource to respective remaining logical volumes for respective remaining compute resources in the distributed storage platform; and executing at least one of the first redistribution plan and the second redistribution plan to reassign data block ownerships on one or more physical memory devices in the storage subsystem, by modifying a mapping of respective logical block addresses in the storage subsystem to respective data block addresses on the one or more physical memory devices; wherein the first redistribution plan is a list of pairs, wherein each pair includes a source logical block address and a target logical block address, wherein the source logical block address is an address of a logical block in an existing logical volume assigned to an existing compute resource and the target logical block address is an address of a logical block in the newly added logical volume; and wherein the second redistribution plan is a list of pairs, wherein each pair includes a source logical block address and a target logical block address, wherein the source logical block address is an address of a logical block in the logical volume for the removed compute resource and the target logical block address is an address of a logical block in a logical volume for a remaining compute resource.

14 . The computer-implemented method of claim 13 , further comprising:
in response to that the newly added compute resource is connected to the storage subsystem, creating an empty logical volume as the newly added logical volume in the storage subsystem; and assigning the empty logical volume to the newly added compute resource.

15 . The computer-implemented method of claim 13 , further comprising:
in response to that the newly added compute resource is connected to the storage subsystem, modifying a mapping of logical block addresses in the global address space of the storage subsystem to the compute resources including the newly added compute resources.

16 . The computer-implemented method of claim 13 , further comprising:
in response to that the removed compute resource is disconnected from the storage subsystem, modifying a mapping of logical block addresses in the global address space of the storage subsystem to the respective remaining compute resources.

17 . (canceled)

18 . (canceled)

Assignees

• IBM

Inventors

• Reyes Sergio
• Twichell Brian Chase

Classifications

• G06F3/064

  Management of blocks · CPC title

• G06F3/0631 Primary

  by allocating resources to storage systems · CPC title

• G06F3/0607

  by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device · CPC title

• G06F3/0689

  Disk arrays, e.g. RAID, JBOD · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title