Startup accelerating method and apparatus

Patent metadata

Field	Value
Publication number	US-10318310-B2
Application number	US-201515309095-A
Country	US
Kind code	B2
Filing date	May 5, 2015
Priority date	May 5, 2014
Publication date	Jun 11, 2019
Grant date	Jun 11, 2019

Abstract

Described is a startup accelerating method and apparatus. The method includes: obtaining, when it is monitored that a login process of an application is started up, pre-fetched data corresponding to a main process of the application, and loading the obtained pre-fetched data into a cache, the pre-fetched data being obtained according to a historical startup procedure for the main process of the application; stopping, when it is monitored that the startup of the login process is completed or it is monitored that the main process of the application is started up, obtaining the pre-fetched data corresponding to the main process of the application; and completing a startup procedure of the main process according to the pre-fetched data loaded in the cache. The apparatus includes: an obtaining module, a stopping module and a startup module.

First claim

The invention claimed is:

1. A startup accelerating method, comprising:
obtaining, when it is monitored that a login process of an application is started up, pre-fetched data corresponding to a main process of the application;
loading the pre-fetched data into a cache, the pre-fetched data being obtained according to a historical startup procedure for the main process of the application;
stopping, when it is monitored that a startup of the login process is completed or it is monitored that the main process of the application is started up, obtaining the pre-fetched data corresponding to the main process of the application; and
completing a startup procedure of the main process according to the pre-fetched data loaded in the cache,
wherein the obtaining pre-fetched data corresponding to a main process of the application, comprises:
obtaining pre-fetched information corresponding to the main process of the application according to an identifier of the login process, wherein the identifier of the login process is a name of the login process;
reading corresponding data from a disk file according to the pre-fetched information; and
determining that the read data is the pre-fetched data corresponding to the main process of the application,

wherein a time period of obtaining the pre-fetched data is equal to a time period of performing the startup procedure of the login process, and
wherein, when a size of the pre-fetched data obtained in the startup procedure of the login process is less than a size of total data corresponding to the pre-fetched information at a time of monitoring that a startup of the login process is completed or at a time of monitoring that the main process of the application is started up, a remaining portion of the total data corresponding to the pre-fetched information is not pre-fetched.

2. The method according to claim 1 , wherein the obtaining pre-fetched information corresponding to the main process of the application according to the identifier of the login process comprises:
obtaining a main process identifier of the application according to the identifier of the login process;
determining a storage path of the pre-fetched information corresponding to the main process according to the main process identifier; and
obtaining the pre-fetched information corresponding to the main process according to the storage path of the pre-fetched information corresponding to the main process.

3. The method according to claim 1 , wherein the reading corresponding data from the disk file according to the pre-fetched information comprises:
parsing the pre-fetched information to obtain parsed pre-fetched information;
forming data information of a specific data structure by the parsed pre-fetched information, the data information comprising a file path, an offset value and a length value of needed data of the main process of the application; and
reading the corresponding data from the disk file according to the file path, the offset value and the length value.

4. The method according to claim 1 , wherein the method further comprises:
calculating a startup progress of the main process according to the size of the total data corresponding to the pre-fetched information and a size of loaded data of the main process in the startup procedure of the main process of the application;
calculating an accelerated time of the main process according to a startup consumed time of the main process, the size of the loaded data of the main process, the size of the pre-fetched data, and the startup progress, the accelerated time indicating a reduction in the startup consumed time that is saved based on the size of the pre-fetched data; and
displaying the startup progress and the accelerated time of the main process.

5. The method according to claim 4 , wherein the accelerated time of the main process is calculated based on the following formula:
T=t/z 1 ×z 2 ×s,
wherein T is the accelerated time of the main process, t is the startup consumed time of the main process, z 1 is the size of the loaded data of the main process, z 2 is the size of the pre-fetched data of the main process, and s is the startup progress.

6. The method according to claim 4 , wherein the startup progress is calculated by dividing the size of the loaded data of the main process by the size of the total data corresponding to the pre-fetched information.

7. The method according to claim 1 , wherein before the obtaining, when it is monitored that the login process of the application is started up, the pre-fetched data corresponding to the main process of the application, the method comprises:
monitoring the historical startup procedure of the main process of the application, and obtaining startup data of the application;
obtaining, according to the startup data, an offset value and a length value of data read by the main process from a magnetic disk in the startup procedure; and
generating pre-fetched information according to a file path of a file at which the startup data is located, the offset value and the length value.

8. The method according to claim 1 , wherein before the obtaining, when it is monitored that the login process of the application is started up, the pre-fetched data corresponding to the main process of the application, the method comprises:
receiving input/output (IO) training data, the IO training data comprising file data and a relative file path of a file at which file data corresponding to the main process is located;
converting, according to an installation path of the main process in a terminal, the relative file path of the file at which the file data corresponding to the main process is located into a corresponding absolute file path;
converting the file data into a corresponding offset value and a corresponding length value; and
generating pre-fetched information according to the corresponding absolute file path, the corresponding offset value and the length value.

9. The method according to claim 1 , wherein the method further comprises, when the size of the pre-fetched data obtained in the startup procedure of the login process is less than the size of the total data corresponding to the pre-fetched information:
reading, from a magnetic disk, the remaining portion of the total data corresponding to the pre-fetched information that has not been pre-fetched, in the startup procedure of the main process; and
completing the startup procedure of the main process according to the pre-fetched data loaded in the cache and the remaining portion read from the magnetic disk in the startup procedure of the main process.

10. A startup accelerating apparatus, comprising:
a processor and a non-transitory storage, wherein the non-transitory storage stores machine readable instructions, and the processor executes the machine readable instructions stored in the non-transitory storage to:
obtain, when it is monitored that a login process of an application is started up, pre-fetched data corresponding to a main process of the application, and load the pre-fetched data into a cache, the pre-fetched data being obtained according to a historical startup procedure for the main process of the application;
stop, when it is monitored that the startup of the login process is completed or it is monitored that the main process of the application is started up, obtaining the pre-fetched data corresponding to the main process of the application; and
complete a startup procedure of the main process according to the pre-fetched data loaded in the cache;
wherein the processor executes the machine readable instructions stored in the non-transitory storage to:
obtain pre-fetched information corresponding to

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Wei Xue
• Jin Qianwen
• Wang Wenqiang
• Li Xuyang
• Gao Kang
• Chen Qiru

Classifications

• G06F9/4451

  User profiles; Roaming · CPC title

• G06F9/4403 Primary

  Processor initialisation · CPC title

• G06F9/4406

  Loading of operating system · CPC title

• G06F9/445 Primary

  Program loading or initiating (bootstrapping G06F9/4401; security arrangements for program loading or initiating G06F21/57) · CPC title

• G06F9/44521

  Dynamic linking or loading; Link editing at or after load time, e.g. Java class loading · CPC title