IP Library › Granted Patent US 12681640
Granted Patent B2
US 12681640 · App. 18/632,800 · Granted Jul 14, 2026

Method and apparatus for determining time series-based asynchronous replicaton cycle time

Inventors: Sandeep Chandrashekhara (Shrewsbury, MA); Ramesh Doddaiah (Westborough, MA); Mohammed Aamir VT (Bangalore, IN); Mohammed Asher (Bangalore, IN)
Assignee: Dell Products L.P.
G06F3/061G06F3/064G06F3/065G06F3/0683
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12681640
App. No.
18/632,800
Granted
Jul 14, 2026
Kind
B2
Abstract

A method comprising: generating a write-on-write distribution for a plurality of data units in a data entity, the write-on-write distribution including a plurality of buckets, each bucket corresponding to a different write-on-write time range, each bucket having a respective bucket size that is based on a count of ones of the plurality of data units whose respective predicted write-on-write times fall within the bucket's write-on-write time range; identifying one or more utilization metrics of a storage system where the data entity is hosted; classifying a representative bucket size of the write-on-write distribution and the one or more utilization metrics with a machine learning model to predict a replication period for the data entity, the representative bucket size being a bucket size within a predetermined distance from a mean bucket size of the write-on-write distribution; and configuring the storage system to replicate the data entity in accordance with the replication period.

Claims (46)

1 . A method comprising:

generating a write-on-write distribution for a plurality of data units in a data entity, the write-on-write distribution including a plurality of buckets, each bucket corresponding to a different write-on-write time range, each bucket having a respective bucket size that is based on a count of ones of the plurality of data units whose respective predicted write-on-write times fall within the bucket's write-on-write time range;

identifying one or more utilization metrics of a storage system where the data entity is hosted;

classifying a representative bucket size of the write-on-write distribution and the one or more utilization metrics with a machine learning model to predict a replication period for the data entity, the representative bucket size being a bucket size within a predetermined distance from a mean bucket size of the write-on-write distribution; and

configuring the storage system to replicate the data entity in accordance with the replication period,

wherein the data entity includes a data volume or a storage group and the replication period of the data entity specifies a duration between consecutive replications of the data entity.

2 . The method of claim 1 , wherein generating the write-on-write distribution includes:

identifying a different respective set of use metrics for each of the plurality of data units; and

classifying the respective set of use metrics for each of the plurality of data units with another machine learning model,

wherein the other machine learning model is configured to yield, at least in part, an indication of a predicted write-on-write time.

3 . The method of claim 2 , wherein the respective set of use metrics for any given one of the plurality of data units includes a plurality of time values, each time value identifying a duration between a different pair of successive writes to the given data unit.

4 . The method of claim 1 , wherein the representative bucket size includes a mean bucket size of the write-on-write distribution.

5 . The method of claim 1 , wherein the one or more utilization metrics of the storage system include one or more of CPU utilization, storage device utilization, network bandwidth utilization, cache write pressure, and memory bandwidth utilization.

6 . The method of claim 1 , wherein the one or more utilization metrics include an amount of bandwidth that is used by the storage system for replicating the data entity.

7 . The method of claim 1 , wherein each of the plurality of subsets of the plurality of data units is a proper subset of the plurality of data units.

8 . The method of claim 1 , wherein each of the plurality of data units includes a different respective track of the data entity.

9 . A system, comprising:

a memory; and

at least one processor that is operatively coupled to the memory, the at least one processor being configured to perform the operations of:

generating a write-on-write distribution for a plurality of data units in a data entity, the write-on-write distribution including a plurality of buckets, each bucket corresponding to a different write-on-write time range, each bucket having a respective bucket size that is based on a count of ones of the plurality of data units whose respective predicted write-on-write times fall within the bucket's write-on-write time range;

identifying one or more utilization metrics of a storage system where the data entity is hosted;

classifying a representative bucket size of the write-on-write distribution and the one or more utilization metrics with a machine learning model to predict a replication period for the data entity, the representative bucket size being a bucket size within a predetermined distance from a mean bucket size of the write-on-write distribution; and

configuring the storage system to replicate the data entity in accordance with the replication period,

wherein the data entity includes a data volume or a storage group and the replication period of the data entity specifies a duration between consecutive replications of the data entity.

10 . The system of claim 9 , wherein generating the write-on-write distribution includes:

identifying a different respective set of use metrics for each of the plurality of data units; and

classifying the respective set of use metrics for each of the plurality of data units with another machine learning model,

wherein the other machine learning model is configured to yield, at least in part, an indication of a predicted write-on-write time.

11 . The system of claim 10 , wherein the respective set of use metrics for any given one of the plurality of data units includes a plurality of time values, each time value identifying a duration between a different pair of successive writes to the given data unit.

12 . The system of claim 9 , wherein the representative bucket size includes a mean bucket size of the write-on-write distribution.

13 . The system of claim 9 , wherein the one or more utilization metrics of the storage system include one or more of CPU utilization, storage device utilization, network bandwidth utilization, cache write pressure, and memory bandwidth utilization.

14 . The system of claim 9 , wherein the one or more utilization metrics include an amount of bandwidth that is used by the storage system for replicating the data entity.

15 . The system of claim 9 , wherein each of the plurality of subsets of the plurality of data units is a proper subset of the plurality of data units.

16 . The system of claim 9 , wherein each of the plurality of data units includes a different respective track of the data entity.

17 . A non-transitory computer-readable medium storing one or more processor-executable instructions, which, when executed by at least one processor, cause the at least one processor to perform the operations of:

generating a write-on-write distribution for a plurality of data units in a data entity, the write-on-write distribution including a plurality of buckets, each bucket corresponding to a different write-on-write time range, each bucket having a respective bucket size that is based on a count of ones of the plurality of data units whose respective predicted write-on-write times fall within the bucket's write-on-write time range;

identifying one or more utilization metrics of a storage system where the data entity is hosted;

classifying a representative bucket size of the write-on-write distribution and the one or more utilization metrics with a machine learning model to predict a replication period for the data entity, the representative bucket size being a bucket size within a predetermined distance from a mean bucket size of the write-on-write distribution; and

configuring the storage system to replicate the data entity in accordance with the replication period,

wherein the data entity includes a data volume or a storage group and the replication period of the data entity specifies a duration between consecutive replications of the data entity.

18 . The non-transitory computer-readable medium of claim 17 , wherein generating the write-on-write distribution includes:

identifying a different respective set of use metrics for each of the plurality of data units; and

classifying the respective set of use metrics for each of the plurality of data units with another machine learning model,

wherein the other machine learning model is configured to yield, at least in part, an indication of a predicted write-on-write time.

19 . The non-transitory computer-readable medium of claim 18 , wherein the respective set of use metrics for any given one of the plurality of data units includes a plurality of time values, each time value identifying a duration between a different pair of successive writes to the given data unit.

20 . The non-transitory computer-readable medium of claim 17 , wherein the representative bucket size includes a mean bucket size of the write-on-write distribution.