# US 12,217,094 B1 · Method of resource adjustment for service cluster, electronic device and storage medium

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

- Number: US 12,217,094 B1
- Document ID: US12217094B1
- Status: Granted
- Kind: B1
- Application number: 18/749,020
- Filed: 2024-06-20
- Published: 2025-02-06
- Granted: 2025-02-04
- Earliest claimed: 2023-07-31
- Assignees: Beijing Volcano Engine Technology Co., Ltd.; National University of Singapore
- Applicants: National University of Singapore; Beijing Volcano Engine Technology Co., Ltd.
- Inventors: Yifan Zhang; Zhanghao Chen; Meng Wang; Guanghui Zhang; Yong Fang; Rui Shi; Tianbai Ma; Yancan Mao
- Web page: https://patentgrep.com/patent/US12217094B1

## CPC

- G06F 9/5038: {considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration (scheduling strategies G06F9/4881 and subgroups)}
- G06F 9/5083: {Techniques for rebalancing the load in a distributed system}
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

The present disclosure provides a method and an apparatus of resource adjustment for a service cluster, an electronic device and a storage medium. The method includes: determining resource adjustment information for the service cluster according to a task quantity of streaming data processing tasks to be processed; determining a core quantity to be adjusted corresponding to central processor units and the second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, the first quantity of the computing nodes that have been created in the service cluster, and a resource utilization rate of central processor units in each computing node that has been created; and executing the resource adjustment operation according to the core quantity to be adjusted and the second quantity, and obtaining a service cluster subject to resource adjusting.

## Claims

### Claim 1

A method of resource adjustment for a service cluster, comprising:
determining resource adjustment information for the service cluster according to a task quantity of streaming data processing tasks to be processed, wherein the resource adjustment information comprises a resource adjustment operation that is required to be executed for processing the streaming data processing tasks and a corresponding parallelism degree to be adjusted, and the parallelism degree to be adjusted is used to indicate a resource quantity that is required to be adjusted by the service cluster;
determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, a first quantity of the computing nodes that have been created in the service cluster, and a resource utilization rate of central processor units in each computing node that has been created; and
executing the resource adjustment operation according to the core quantity to be adjusted and the second quantity, and obtaining a service cluster subject to resource adjusting, wherein the service cluster subject to resource adjusting is used to process streaming data processing tasks under the task quantity.

### Claim 2 (depends on 1)

The method according to claim 1, wherein, in a case where the resource adjustment operation comprises a resource scaling-out operation, the core quantity to be adjusted comprises a core quantity to be increased, and the computing nodes to be adjusted comprise computing nodes to be increased;
the determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, the first quantity of the computing nodes that have been created in the service cluster, and the resource utilization rate of the central processor units in each computing node that has been created, comprises:
determining the core quantity to be adjusted of the central processor units, according to the resource utilization rate, an initial core quantity of the central processor units and a node quantity of resource nodes in the computing nodes that have been created;
determining a conversion parallelism degree corresponding to the core quantity to be increased according to the core quantity to be increased and the first quantity; the conversion parallelism degree being used to characterize a parallelism degree that is equivalently increased based on the total original parallelism degree corresponding to the service cluster, after respectively adding the core quantity to be increased to the central processor units in each computing node that has been created; and
in a case where the conversion parallelism degree is less than the parallelism degree to be adjusted, determining a second quantity of the computing nodes to be increased according to the conversion parallelism degree, the parallelism degree to be adjusted and the node quantity.

### Claim 3 (depends on 2)

The method according to claim 2, wherein the determining the core quantity to be adjusted of the central processor units, according to the resource utilization rate, the initial core quantity of the central processor units and the node quantity of the resource nodes in the computing nodes that have been created, comprises:
in a case where the initial core quantity is less than a preset core quantity threshold value, determining whether a target resource utilization rate that is larger than a first preset threshold value exists in each resource utilization rate;
in response to the target resource utilization rate that is larger than the first preset threshold value existing in each resource utilization rate, determining a target core quantity according to the target resource utilization rate, the initial core quantity and a second preset threshold value; and
determining the core quantity to be increased according to a minimum value of the target core quantity, the node quantity and the preset core quantity threshold value, and the initial core quantity.

### Claim 4 (depends on 2)

The method according to claim 2, wherein the determining a conversion parallelism degree corresponding to the core quantity to be increased according to the core quantity to be increased and the first quantity, comprises:
determining a total increased core quantity according to the core quantity to be adjusted and the first quantity; and
determining the conversion parallelism degree according to the node quantity and a ratio of the total increased core quantity to the initial core quantity.

### Claim 5 (depends on 2)

The method according to claim 2, wherein the determining a second quantity of the computing nodes to be increased according to the conversion parallelism degree, the parallelism degree to be adjusted and the node quantity, comprises:
determining a difference parallelism degree between the parallelism degree to be adjusted and the conversion parallelism degree; and
in a case where the difference parallelism degree is larger than a target threshold value, determining the second quantity of the computing nodes to be increased according to a ratio of the difference parallelism degree to the node quantity.

### Claim 6 (depends on 1)

The method according to claim 1, wherein, in a case where the resource adjustment operation comprises a resource reduction operation, the core quantity to be adjusted comprises a core quantity to be reduced, and the computing nodes to be adjusted comprise computing nodes to be reduced;
the determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, the first quantity of the computing nodes that have been created in the service cluster, and the resource utilization rate of the central processor units in each computing node that has been created, comprises:
in a case where the first quantity is larger than a preset quantity, determining a second quantity of the computing nodes to be reduced, according to the parallelism degree to be adjusted and a node quantity of resource nodes in the computing nodes that have been created;
in a case where a difference value between the second quantity of the computing nodes to be reduced and the first quantity is equal to the preset quantity, determining whether the resource utilization rate of the computing nodes that have been created of the preset quantity is less than a third preset threshold value; and
in response to the resource utilization rate of the computing nodes that have been created of the preset quantity being less than the third preset threshold value, determining the core quantity to be reduced of the central processor units according to the resource utilization rate, the initial core quantity of the central processor units and a second preset threshold value.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the determining resource adjustment information for the service cluster according to a task quantity of streaming data processing tasks to be processed, comprises:
determining a target parallelism degree corresponding to the streaming data processing tasks according to the task quantity;
determining a total original parallelism degree corresponding to the service cluster, according to the first quantity and a node quantity of resource nodes in the computing nodes that have been created; and
determining the resource adjustment operation and the parallelism degree to be adjusted, according to a difference value between the target parallelism degree and the total original parallelism degree.

### Claim 8

An electronic device, comprising:
at least one processor and a memory, wherein the memory stores machine-readable instructions executed by the at least one processor, the at least one processor is configured to execute the machine-readable instructions stored in the memory, and when the machine-readable instructions are executed by the at least one processor, the at least one processor is caused to execute:
determining resource adjustment information for the service cluster according to a task quantity of streaming data processing tasks to be processed, wherein the resource adjustment information comprises a resource adjustment operation that is required to be executed for processing the streaming data processing tasks and a corresponding parallelism degree to be adjusted; the parallelism degree to be adjusted is used to indicate a resource quantity that is required to be adjusted by the service cluster;
determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, a first quantity of the computing nodes that have been created in the service cluster, and a resource utilization rate of central processor units in each computing node that has been created; and
executing the resource adjustment operation according to the core quantity to be adjusted and the second quantity, and obtaining a service cluster subject to resource adjusting, wherein the service cluster subject to resource adjusting is used to process streaming data processing tasks under the task quantity.

### Claim 9 (depends on 8)

The electronic device according to claim 8, wherein, in a case where the resource adjustment operation comprises a resource scaling-out operation, the core quantity to be adjusted comprises a core quantity to be increased, and the computing nodes to be adjusted comprise computing nodes to be increased;
the determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, the first quantity of the computing nodes that have been created in the service cluster, and the resource utilization rate of the central processor units in each computing node that has been created, comprises:
determining the core quantity to be adjusted of the central processor units, according to the resource utilization rate, an initial core quantity of the central processor units and a node quantity of resource nodes in the computing nodes that have been created;
determining a conversion parallelism degree corresponding to the core quantity to be increased according to the core quantity to be increased and the first quantity; the conversion parallelism degree being used to characterize a parallelism degree that is equivalently increased based on the total original parallelism degree corresponding to the service cluster, after respectively adding the core quantity to be increased to the central processor units in each computing node that has been created; and
in a case where the conversion parallelism degree is less than the parallelism degree to be adjusted, determining a second quantity of the computing nodes to be increased according to the conversion parallelism degree, the parallelism degree to be adjusted and the node quantity.

### Claim 10 (depends on 9)

The electronic device according to claim 9, wherein the determining the core quantity to be adjusted of the central processor units, according to the resource utilization rate, the initial core quantity of the central processor units and the node quantity of the resource nodes in the computing nodes that have been created, comprises:
in a case where the initial core quantity is less than a preset core quantity threshold value, determining whether a target resource utilization rate that is larger than a first preset threshold value exists in each resource utilization rate;
in response to the target resource utilization rate that is larger than the first preset threshold value existing in each resource utilization rate, determining a target core quantity according to the target resource utilization rate, the initial core quantity and a second preset threshold value; and
determining the core quantity to be increased according to a minimum value of the target core quantity, the node quantity and the preset core quantity threshold value, and the initial core quantity.

### Claim 11 (depends on 9)

The electronic device according to claim 9, wherein the determining a conversion parallelism degree corresponding to the core quantity to be increased according to the core quantity to be increased and the first quantity, comprises:
determining a total increased core quantity according to the core quantity to be adjusted and the first quantity; and
determining the conversion parallelism degree according to the node quantity and a ratio of the total increased core quantity to the initial core quantity.

### Claim 12 (depends on 9)

The electronic device according to claim 9, wherein the determining a second quantity of the computing nodes to be increased according to the conversion parallelism degree, the parallelism degree to be adjusted and the node quantity, comprises:
determining a difference parallelism degree between the parallelism degree to be adjusted and the conversion parallelism degree; and
in a case where the difference parallelism degree is larger than a target threshold value, determining the second quantity of the computing nodes to be increased according to a ratio of the difference parallelism degree to the node quantity.

### Claim 13 (depends on 8)

The electronic device according to claim 8, wherein, in a case where the resource adjustment operation comprises a resource reduction operation, the core quantity to be adjusted comprises a core quantity to be reduced, and the computing nodes to be adjusted comprise computing nodes to be reduced;
the determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, the first quantity of the computing nodes that have been created in the service cluster, and the resource utilization rate of the central processor units in each computing node that has been created, comprises:
in a case where the first quantity is larger than a preset quantity, determining a second quantity of the computing nodes to be reduced, according to the parallelism degree to be adjusted and a node quantity of resource nodes in the computing nodes that have been created;
in a case where a difference value between the second quantity of the computing nodes to be reduced and the first quantity is equal to the preset quantity, determining whether the resource utilization rate of the computing nodes that have been created of the preset quantity is less than a third preset threshold value; and
in response to the resource utilization rate of the computing nodes that have been created of the preset quantity being less than the third preset threshold value, determining the core quantity to be reduced of the central processor units according to the resource utilization rate, the initial core quantity of the central processor units and a second preset threshold value.

### Claim 14 (depends on 8)

The electronic device according to claim 8, wherein the determining resource adjustment information for the service cluster according to a task quantity of streaming data processing tasks to be processed, comprises:
determining a target parallelism degree corresponding to the streaming data processing tasks according to the task quantity;
determining a total original parallelism degree corresponding to the service cluster, according to the first quantity and a node quantity of resource nodes in the computing nodes that have been created; and
determining the resource adjustment operation and the parallelism degree to be adjusted, according to a difference value between the target parallelism degree and the total original parallelism degree.

### Claim 15

A non-transient computer-readable storage medium, wherein the non-transient computer-readable storage medium stores computer programs, and when the computer programs are run by an electronic device, the electronic device is caused to execute:
determining resource adjustment information for the service cluster according to a task quantity of streaming data processing tasks to be processed, wherein the resource adjustment information comprises a resource adjustment operation that is required to be executed for processing the streaming data processing tasks and a corresponding parallelism degree to be adjusted; the parallelism degree to be adjusted is used to indicate a resource quantity that is required to be adjusted by the service cluster;
determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, a first quantity of the computing nodes that have been created in the service cluster, and a resource utilization rate of central processor units in each computing node that has been created; and
executing the resource adjustment operation according to the core quantity to be adjusted and the second quantity, and obtaining a service cluster subject to resource adjusting, wherein the service cluster subject to resource adjusting is used to process streaming data processing tasks under the task quantity.

### Claim 16 (depends on 15)

The non-transient computer-readable storage medium according to claim 15, wherein in a case where the resource adjustment operation comprises a resource scaling-out operation, the core quantity to be adjusted comprises a core quantity to be increased, and the computing nodes to be adjusted comprise computing nodes to be increased;
the determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, the first quantity of the computing nodes that have been created in the service cluster, and the resource utilization rate of the central processor units in each computing node that has been created, comprises:
determining the core quantity to be adjusted of the central processor units, according to the resource utilization rate, an initial core quantity of the central processor units and a node quantity of resource nodes in the computing nodes that have been created;
determining a conversion parallelism degree corresponding to the core quantity to be increased according to the core quantity to be increased and the first quantity; the conversion parallelism degree being used to characterize a parallelism degree that is equivalently increased based on the total original parallelism degree corresponding to the service cluster, after respectively adding the core quantity to be increased to the central processor units in each computing node that has been created; and
in a case where the conversion parallelism degree is less than the parallelism degree to be adjusted, determining a second quantity of the computing nodes to be increased according to the conversion parallelism degree, the parallelism degree to be adjusted and the node quantity.

### Claim 17 (depends on 16)

The non-transient computer-readable storage medium according to claim 16, wherein the determining the core quantity to be adjusted of the central processor units, according to the resource utilization rate, the initial core quantity of the central processor units and the node quantity of the resource nodes in the computing nodes that have been created, comprises:
in a case where the initial core quantity is less than a preset core quantity threshold value, determining whether a target resource utilization rate that is larger than a first preset threshold value exists in each resource utilization rate;
in response to the target resource utilization rate that is larger than the first preset threshold value existing in each resource utilization rate, determining a target core quantity according to the target resource utilization rate, the initial core quantity and a second preset threshold value; and
determining the core quantity to be increased according to a minimum value of the target core quantity, the node quantity and the preset core quantity threshold value, and the initial core quantity.

### Claim 18 (depends on 16)

The non-transient computer-readable storage medium according to claim 16, wherein the determining a conversion parallelism degree corresponding to the core quantity to be increased according to the core quantity to be increased and the first quantity, comprises:
determining a total increased core quantity according to the core quantity to be adjusted and the first quantity; and
determining the conversion parallelism degree according to the node quantity and a ratio of the total increased core quantity to the initial core quantity.

### Claim 19 (depends on 16)

The non-transient computer-readable storage medium according to claim 16, wherein the determining a second quantity of the computing nodes to be increased according to the conversion parallelism degree, the parallelism degree to be adjusted and the node quantity, comprises:
determining a difference parallelism degree between the parallelism degree to be adjusted and the conversion parallelism degree; and
in a case where the difference parallelism degree is larger than a target threshold value, determining the second quantity of the computing nodes to be increased according to a ratio of the difference parallelism degree to the node quantity.

### Claim 20 (depends on 15)

The non-transient computer-readable storage medium according to claim 15, wherein in a case where the resource adjustment operation comprises a resource reduction operation, the core quantity to be adjusted comprises a core quantity to be reduced, and the computing nodes to be adjusted comprise computing nodes to be reduced;
the determining a core quantity to be adjusted corresponding to central processor units and a second quantity of computing nodes to be adjusted, according to the parallelism degree to be adjusted, the first quantity of the computing nodes that have been created in the service cluster, and the resource utilization rate of the central processor units in each computing node that has been created, comprises:
in a case where the first quantity is larger than a preset quantity, determining a second quantity of the computing nodes to be reduced, according to the parallelism degree to be adjusted and a node quantity of resource nodes in the computing nodes that have been created;
in a case where a difference value between the second quantity of the computing nodes to be reduced and the first quantity is equal to the preset quantity, determining whether the resource utilization rate of the computing nodes that have been created of the preset quantity is less than a third preset threshold value; and
in response to the resource utilization rate of the computing nodes that have been created of the preset quantity being less than the third preset threshold value, determining the core quantity to be reduced of the central processor units according to the resource utilization rate, the initial core quantity of the central processor units and a second preset threshold value.