Granted Patent B2

US 12,487,865 · App. 17/445,676 · Granted Dec 2, 2025

Inventors: Gary W. Grube (Barrington Hills, IL); Timothy W. Markison (Mesa, AZ)

Assignee: Pure Storage, Inc.

G06F9/5083G06F9/5066G06F16/27G06F16/273G06F16/275G06F16/278H04L67/10H04L67/1097

Abstract

Claims (49)

1 . A method for execution by one or more computing devices of a storage network, the method comprising:

selecting a number of storage and execution units from a plurality of storage and execution units of the storage network for a task based on a capability level associated with each storage and execution unit of the plurality of storage and execution units to produce a selected subset of storage and execution units, wherein the task is associated with data;

determining processing parameters for the data based on the number of the selected subset of storage and execution units to be utilized in processing the data;

determining task partitioning of the task based on the number of the selected subset of storage and execution units and the determined processing parameters;

processing the data in accordance with the determined processing parameters to produce slice groupings;

partitioning the task based on the task partitioning to produce partial tasks; and

sending the slice groupings and corresponding partial tasks to the selected subset of storage and execution units for execution.

2 . The method of claim 1 further comprises:

receiving the data and the task.

3 . The method of claim 1 , wherein the selecting is based on one or more of:

a size of the data;

the task;

the content of the data;

a predetermined number; and

a number of available storage and execution units.

4 . The method of claim 1 , wherein the selecting is based on an estimated distributed computing loading level.

5 . The method of claim 1 , wherein the selecting is based on a storage and execution unit capability indicator.

6 . The method of claim 1 , wherein the selecting is based on a storage and execution unit performance indicator.

7 . The method of claim 1 , wherein the selecting is based on a storage and execution unit availability level indicator.

8 . The method of claim 1 , wherein the selecting is based on a storage and execution unit threshold computing capability indicator.

9 . The method of claim 1 , wherein the selecting is based on a task schedule.

10 . The method of claim 1 , wherein the determining the task partitioning further comprises:

determining partial task execution ordering for the selected subset of the storage and execution units.

11 . The method of claim 10 further comprises:

sending the slice grouping and corresponding partial task to the selected subset of storage and execution units in accordance with the partial task execution ordering.

12 . The method of claim 10 , wherein the partial task execution ordering includes an indication of which slice of a corresponding slice grouping that a corresponding storage and execution unit of the selected subset of storage and execution units is to process first.

13 . The method of claim 1 , wherein the processing the data in accordance with the processing parameters to produce slice groupings includes:

arranging the data into a plurality of chunksets based on a chunk size, a data size of the data, and the number of the selected subset of storage and execution units;

generating a chunkset data matrix based on a first chunkset of the plurality of chunksets; and

matrix multiplying the chunkset data matrix with a generator matrix to produce a chunkset slice matrix, wherein each row of the chunkset slice matrix is a slice grouping of the slice groupings.

14 . The method of claim 13 , wherein the generating the chunkset data matrix comprises:

determining a decode threshold number for the slice grouping;

arranging the first chunkset into the chunkset data matrix, wherein a number of rows of the chunkset data matrix corresponds to the decode threshold number.

15 . The method of claim 13 further comprises:

generating the generator matrix to include a decode threshold number columns and a pillar width number of rows, wherein the pillar width number corresponds to the number of the selected subset of storage and execution units.

16 . The method of claim 1 , wherein the processing parameters include data partitioning information that includes a number of data partitions, size of each data partition of the number of data partitions and organization of the data partitions.

17 . The method of claim 1 , wherein the processing parameters include slice grouping information regarding arrangement of encoded data slices into groups to produce the slice groupings.

18 . The method of claim 1 , wherein the processing parameters include error encoding parameters.

19 . The method of claim 18 , wherein the error encoding parameters include a pillar width number and a decode threshold number.

20 . A computing device of a storage network, the computing device comprises:

memory;

an interface; and

a processing module operably coupled to the memory and the interface, wherein the processing module is operable to:

select a number of storage and execution units from a plurality of storage and execution units for a task based on a capability level associated with each storage and execution unit of the storage and execution units to produce a selected subset of storage and execution units, wherein the task is associated with data;

determine processing parameters for the data based on the number of the selected subset of storage and execution units of the storage network to be utilized in processing the data;

determine task partitioning of the task based on the number of the selected subset of storage and execution units and the determined processing parameters;

process the data in accordance with the determined processing parameters to produce slice groupings;

partition the task based on the task partitioning to produce partial tasks; and

send, via the interface, the slice groupings and corresponding partial tasks to the selected subset of storage and execution units for execution.

Assignments (1)

Continuity (5)

References Cited (127)