# US 2020/0401947 A1 · Workload tenure prediction for capacity planning

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2020/0401947 A1
- Document ID: US20200401947A1
- Status: Published application
- Kind: A1
- Application number: 16/444,190
- Filed: 2019-06-18
- Published: 2020-12-24
- Earliest claimed: 2019-06-18
- Assignees: VMware, Inc.
- Applicants: VMware, Inc.
- Inventors: Chandrashekhar Jha; Ritesh Jha; Yash Bhatnagar; Rajat Garg; Rachil Chandran
- Web page: https://patentgrep.com/patent/US20200401947A1

## CPC

- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06F 9/45558: {Hypervisor-specific management and integration aspects}
- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06F 2009/4557: {Distribution of virtual machine instances; Migration and load balancing}
- G06F 2209/5019: Workload prediction
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Disclosed are various embodiments for automating the prediction of workload tenures in datacenter environments. In some embodiments, parameters are identified for a plurality of workloads of a software defined data center. A machine learning model is trained to determine a predicted tenure based on parameters of the workloads. A workload for the software defined data center is configured to include at least one workload parameter. The workload is processed using the trained machine learning model to determine the predicted tenure. An input to the machine learning model includes the at least one workload parameter.

## Claims

### Claim 1

A system comprising:
at least one computing device comprising at least one processor and at least one data store;
machine readable instructions stored in the at least one data store, wherein the instructions, when executed by the at least one processor, cause the at least one computing device to at least:
identify workload data for a plurality of workloads of a software defined data center, the workload data for a respective workload comprising: a measured tenure, and at least one identified workload parameter;
train a machine learning model to determine a predicted tenure based on the measured tenure and the at least one identified workload parameter;
receive a user input to configure a workload for the software defined data center, the workload comprising at least one workload parameter;
process the workload using the machine learning model to determine the predicted tenure for the workload, an input to the machine learning model comprising the at least one workload parameter; and
render, within a user interface, a demand graph for the software defined data center, the demand graph being generated based on the predicted tenure for the workload.

### Claim 2 (depends on 1)

The system of claim 1, wherein the at least one workload parameter comprises an implementation time, and the demand graph is generated based on: the predicted tenure for the workload, and the implementation time.

### Claim 3 (depends on 1)

The system of claim 1, wherein the machine learning model comprises a support vector machine (SVM) regression model.

### Claim 4 (depends on 1)

The system of claim 1, wherein the instructions, when executed by the at least one processor, cause the at least one computing device to at least:
train a plurality of machine learning models to determine the predicted tenure based on the measured tenure and the at least one identified workload parameter;
determine that the machine learning model comprises a minimum error among the plurality of machine learning models; and
select the machine learning model to process the workload.

### Claim 5 (depends on 1)

The system of claim 1, wherein the at least one identified workload parameter comprises at least one identified software parameter.

### Claim 6 (depends on 5)

The system of claim 5, wherein the at least one identified software parameter comprises: an operation system identifier, a fault tolerance indication, and a thin provisioning indication.

### Claim 7 (depends on 1)

The system of claim 1, wherein the at least one identified workload parameter comprises at least one of: a number of virtual central processing units (vCPUs) for the respective workload, a memory capacity for the respective workload, a memory reservation for the respective workload, a number of network cards for the respective workload, and a central processing unit (CPU) reservation for the respective workload.

### Claim 8

A method comprising:
identifying workload data for a plurality of workloads of a software defined data center, the workload data for a respective workload comprising: a measured tenure, and at least one identified workload parameter;
training a machine learning model to determine a predicted tenure based on the measured tenure and the at least one identified workload parameter;
receiving a user input to configure a workload for the software defined data center, the workload comprising at least one workload parameter;
processing the workload using the machine learning model to determine the predicted tenure for the workload, an input to the machine learning model comprising the at least one workload parameter; and
rendering, within a user interface, a demand graph for the software defined data center, the demand graph being generated based on the predicted tenure for the workload.

### Claim 9 (depends on 8)

The method of claim 8, wherein the at least one workload parameter comprises an implementation time, and the demand graph is generated based on: the predicted tenure for the workload, and the implementation time.

### Claim 10 (depends on 8)

The method of claim 8, wherein the machine learning model comprises a support vector machine (SVM) regression model.

### Claim 11 (depends on 8)

The method of claim 8, further comprising:
training a plurality of machine learning models to determine the predicted tenure based on the measured tenure and the at least one identified workload parameter;
determining that the machine learning model comprises a minimum error among the plurality of machine learning models; and
selecting the machine learning model to process the workload.

### Claim 12 (depends on 8)

The method of claim 8, wherein the at least one identified workload parameter comprises at least one identified software parameter.

### Claim 13 (depends on 12)

The method of claim 12, wherein the at least one identified software parameter comprises: an operation system identifier, a fault tolerance indication, and a thin provisioning indication.

### Claim 14 (depends on 8)

The method of claim 8, wherein the at least one identified workload parameter comprises at least one of: a number of virtual central processing units (vCPUs) for the respective workload, a memory capacity for the respective workload, a memory reservation for the respective workload, a number of network cards for the respective workload, and a central processing unit (CPU) reservation for the respective workload.

### Claim 15

A non-transitory computer-readable medium comprising machine readable instructions, wherein the instructions, when executed by at least one processor, cause at least one computing device to at least:
identify workload data for a plurality of workloads of a software defined data center, the workload data for a respective workload comprising: a measured tenure, and at least one identified workload parameter;
train a machine learning model to determine a predicted tenure based on the measured tenure and the at least one identified workload parameter;
receive a user input to configure a workload for the software defined data center, the workload comprising at least one workload parameter;
process the workload using the machine learning model to determine the predicted tenure for the workload, an input to the machine learning model comprising the at least one workload parameter; and
render, within a user interface, a demand graph for the software defined data center, the demand graph being generated based on the predicted tenure for the workload.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the at least one workload parameter comprises an implementation time, and the demand graph is generated based on: the predicted tenure for the workload, and the implementation time.

### Claim 17 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the machine learning model comprises a support vector machine (SVM) regression model.

### Claim 18 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the instructions, when executed by the at least one processor, cause the at least one computing device to at least:
train a plurality of machine learning models to determine the predicted tenure based on the measured tenure and the at least one identified workload parameter;
determine that the machine learning model comprises a minimum error among the plurality of machine learning models; and
select the machine learning model to process the workload.

### Claim 19 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the at least one identified workload parameter comprises at least one identified software parameter.

### Claim 20 (depends on 19)

The non-transitory computer-readable medium of claim 19, wherein the at least one identified software parameter comprises: an operation system identifier, a fault tolerance indication, and a thin provisioning indication.
