Capacity expansion method and apparatus

Patent metadata

Field	Value
Publication number	US-11216310-B2
Application number	US-201916523028-A
Country	US
Kind code	B2
Filing date	Jul 26, 2019
Priority date	Jan 26, 2017
Publication date	Jan 4, 2022
Grant date	Jan 4, 2022

Abstract

A capacity expansion method includes obtaining a measured workload of a service of an application, obtaining an application model of the application, and obtaining a measured workload of each upper-level service of the service; determining a predicted workload of the service based on the measured workload of the service, determining the measured workload of each upper-level service of the first service, and determining a first workload ratio corresponding to a first calling relationship; and determining a predicted workload of each lower-level service based on the predicted workload of the service and determining a second workload ratio corresponding to a second calling relationship.

First claim

What is claimed is:

1. A capacity expansion method, comprising:
obtaining a first measured workload of an application and an application model of the application, wherein the application comprises services, and wherein the first measured workload is of a first service of any of the services, wherein the application model comprises a calling relationship between the services and a workload ratio, wherein the workload ratio corresponds to each of the calling relationships;
determining upper-level services of the first service based on a first calling relationship between the first service and each upper-level service, wherein the application model comprises the first calling relationship;
determining lower-level services of the first service based on a second calling relationship between the first service and each of the lower-level services of the first service, wherein the application model comprises the second calling relationship;
obtaining a second measured workload of each of the upper-level services;
determining a predicted workload of the first service based on the first measured workload, the second measured workload and a first workload ratio that corresponds to the first calling relationship;
determining a predicted workload of each of the lower-level services based on the predicted workload of the first service and a second workload ratio that corresponds to the second calling relationship; and
performing capacity expansion on each target service based on a predicted workload of each of the target services, wherein the target services comprise the first service and each of the lower-level services.

2. The capacity expansion method of claim 1 , further comprising:
obtaining a service interface description file of each of the services and a configuration file of each of the services, wherein the service interface description file comprises a name of each of the services, and wherein the configuration file comprises a calling relationship between each of the services and a lower-level service of each of the services;
determining the calling relationship between the services based on the calling relationship between each of the services and the lower-level service;
obtaining a workload history of each of the services based on the name of each of the services;
determining the workload ratio that corresponds to each calling relationship based on the workload history and the calling relationship between the services; and
generating the application model based on the calling relationship between the services, and the workload ratio.

3. The capacity expansion method of claim 1 , further comprising:
obtaining a workload history of each of the services based on a name of each of the services;
updating the workload ratio based on the workload history and each of the calling relationships; and
updating the application model based on each of the calling relationships and an updated workload ratio that corresponds to each of the calling relationships.

4. The capacity expansion method of claim 1 , further comprising:
obtaining each of a service interface description file of a third service, a configuration file of the third service, and an updated configuration file of each fourth service in response to the third service being added to an updated application, wherein the service interface description file of the third service comprises a name of the third service, wherein the configuration file of the third service comprises a third calling relationship between the third service and each fifth service, wherein the updated configuration file of each of the fourth services comprises a fourth calling relationship between each of the fourth services and the third service, wherein a fourth service of each of the fourth services is an upper-level service of the third service, and wherein a fifth service of each of the fifth services is a lower-level service of the third service;
updating the calling relationship between the services based on the application model, the third calling relationship, and the fourth calling relationship to obtain updated calling relationships;
obtaining a workload history of each service of the updated application based on a name of each of the services of the updated application;
determining a fourth workload ratio that corresponds to each of the updated calling relationships based on the workload history of each of the services of the updated application and an updated calling relationship between the services; and
updating the application model of the application based on the updated calling relationships and the fourth workload ratio.

5. The capacity expansion method of claim 1 , further comprising:
obtaining an updated configuration file of each seventh service in response to a sixth service being deleted from an updated application, wherein before the sixth service is deleted from the updated application, the seventh service comprises an upper-level service of the sixth service, and wherein after the sixth service is deleted from the application, the updated configuration file of each seventh service comprises a fifth calling relationship between each of the seventh services and an eighth service, and wherein the eighth service is a lower-level service of each of the seventh services;
updating the calling relationship between the services based on the application model and the fifth calling relationship to obtain updated calling relationships;
obtaining a workload history of each service of the updated application based on a name of each of the services of the updated application;
determining a third workload ratio that corresponds to each of the updated calling relationships between the services based on the workload history of each of the services of the updated application and an updated calling relationship between the services; and
updating the application model of the application based on the updated calling relationships between the services, and the third workload ratio.

6. The capacity expansion method of claim 1 , wherein determining the predicted workload of the first service comprises determining the predicted workload of the first service according to a preset formula comprising f(v i )=max(d(v i ),Σ k∈K f(k)*e ki ), wherein V indicates a set of all the services of the application, wherein K indicates a set of upper-level services k of a service i of the application, wherein K∈V, v i indicates the service i, wherein d(v i ) indicates a measured workload of the service i, wherein f(k) indicates a measured workload of the upper-level service k of the service i, wherein e ki indicates a workload ratio between the service k and the service i, and wherein the service i is any one of the services.

7. The capacity expansion method of claim 1 , wherein performing the capacity expansion on each of the target services comprises:
determining a first quantity of instances to be expanded for each of the target services based on the predicted workload of each of the target services and a prestored correspondence between a workload of each of the target services and a quantity of instances; and
performing capacity expansion on each of the target services based on the first quantity of instances.

8. A capacity expansion apparatus, comprising
a processor, and
a memory coupled to the processor and configured to store a plurality of instructions that, when executed by the processor, causes the processor to:
obtain a first measured workload of a first service of an application, wherein the capacity expansion apparatus comprises the application;
generate an application model of the application, wherein the application model comprises a calling relationship between services of the application and a workload rati

Assignees

• Huawei Tech Co Ltd

Inventors

• Zhuo Donghui
• Xu Jun
• Shan Haijun

Classifications

• G06F9/5061 Primary

  Partitioning or combining of resources · CPC title

• G06F9/5027 Primary

  the resource being a machine, e.g. CPUs, Servers, Terminals · CPC title

• H04L67/51 Primary

  Discovery or management thereof, e.g. service location protocol [SLP] or web services · CPC title

• G06F9/5077

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F2209/5019

  Workload prediction · CPC title