Android application installation optimization method and system, mobile terminal and storage medium

Abstract

The application provides an Android application installation optimization method and system, a mobile terminal and a storage medium, and the method steps comprise the following steps: S1, receiving an application package to be installed from different application installation sources, and analyzing identification information of the application package; S2, searching external configuration file resources according to the identification information, screening out an optimized configuration file, and extracting a preset benchmark configuration file in the package to obtain a candidate configuration file; S3, calculating the adaptation degree of the candidate configuration file and a target device based on multi-dimensional portrait information, determining an adapted target configuration file, and performing pre-evaluation of an optimization effect; S4, configuring an installation mode based on the processing results of steps S2 to S3, generating a metadata file and jointly submitting the metadata file and the application package for installation; and S5, monitoring the generation of a runtime configuration file of an installed application, obtaining the configuration file, appending multi-dimensional portrait information, comparing the configuration file with existing configuration files in the cloud, and uploading the configuration file when the difference meets the condition, so as to be reused in subsequent installation. In this way, the application installation of different sources is optimized.

Description

Technical Field

[0001] This invention relates to the field of mobile terminal software installation and operation performance optimization technology, and in particular to a method, system, mobile terminal and storage medium for modifying the Android application installation system to adapt to different application markets, enterprise distribution platforms and local APK sources, and uniformly executing the collection, uploading, matching and distribution of application runtime profiles and differentiated installation. Background Technology

[0002] Android applications are typically distributed as APK installation packages, with application code primarily existing in DEX (DalvikExecutable) bytecode form. After an application is installed on a device, the system does not immediately compile all code into optimal native machine code for all scenarios. Instead, the Android Runtime (ART) dynamically selects and gradually optimizes the code execution method based on factors such as installation timing, device resources, actual user behavior, and system policies. Therefore, the application's cold start speed, first-screen display speed, page switching efficiency, and list scrolling smoothness are closely related to the Android Runtime's (ART) optimization of frequently used classes and methods.

[0003] In the Android system, the Android Runtime (ART) is the current application runtime environment responsible for runtime tasks such as application code loading, interpretation, just-in-time (JIT) compilation, pre-compilation, and garbage collection. For the same piece of application code, ART can process it using different methods, including interpretation, JIT (Just-In-Time) compilation, or AOT (Ahead-Of-Time) compilation. Generally speaking, interpretation is less efficient, AOT compilation results in higher efficiency but increases installation time and storage space usage, while JIT falls between the two, allowing for incremental optimization of frequently accessed code during application runtime.

[0004] In the Android system, dexopt is a processing flow that verifies, optimizes, and compiles DEX bytecode. This flow typically employs different compile filter modes depending on the triggering time, such as during application installation, device idle charging, and after system OTA updates. The `speed-profile` mode prioritizes compiling hotspot code based on the application profile; therefore, if a usable profile is unavailable, the optimization benefits will be significantly limited, and in some scenarios, it may degenerate into a more conservative verify mode or a low-optimization (run-from-apk) mode.

[0005] To strike a balance between installation time, storage usage, and runtime performance, major Android versions have continuously adjusted application compilation optimization strategies. For example, in Android 9 and later, the system's support for Profile-Guided Optimization is more mature. The system can combine the Baseline Profile provided during installation with the Cloud Profile obtained later (mainly aggregated and distributed by Google Play) to perform more targeted pre-compilation of application critical paths. As major Android versions have evolved, the Android system has gradually shifted from a heavier full pre-compilation mode to a more localized optimization mode that relies more on configuration files (Profiles) for guidance.

[0006] Currently, common optimization techniques in existing technologies mainly include the following categories:

[0007] 1) Application developers pre-configure a baseline profile during the build phase to improve critical path execution efficiency after the application's first installation;

[0008] 2) The system automatically accumulates runtime configuration files (RuntimeProfile) after the application has been running for a long time, and then combines them with the background dexopt for compilation optimization;

[0009] 3) The app store or distribution channel (mainly Google Play) optimizes the installation process under specific conditions in conjunction with profile data.

[0010] In terms of application sources, they generally come from app stores such as Google Play and Tencent App Store, as well as the app stores of major ROM manufacturers; some are self-updated applications released by application manufacturers themselves, and applications developed and downloaded by users. Regardless of the source of the application, the actual installation and execution still rely on system installers or specific installation components to install the system.

[0011] However, most existing installation systems only perform "normal installation" and lack the ability to identify, manage and utilize application profile optimization data. As a result, even if an application market, enterprise platform or local source has corresponding profile resources, they cannot be stably accumulated and converted into performance benefits during the installation phase.

[0012] On the other hand, the formation of runtime profiles depends on actual usage. If the installation system cannot directly provide the ability to collect and export application profiles, then even if applications from different sources are frequently used on the terminal, it will be difficult to accumulate hot execution data into reusable and optimized assets.

[0013] Therefore, the existing Android installation system lacks a mechanism that can adapt to multiple app stores and multiple APK sources, and uniformly complete the collection, uploading and optimization of application runtime profiles during the installation stage. This makes it difficult to accumulate and reuse application profiles, and it is difficult to standardize the installation behavior, resulting in unstable performance improvement after the first installation of the application. Summary of the Invention

[0014] Therefore, the main objective of this invention is to provide an Android application installation optimization method and system, mobile terminal, and storage medium to solve the problems mentioned in the background art.

[0015] To achieve the above objectives, according to one aspect of the present invention, an Android application installation optimization method is provided, the steps of which include:

[0016] Step S1 receives application packages to be installed from different application installation sources and parses the identification information of the application packages;

[0017] Step S2 retrieves external configuration file resources based on the identification information, filters out optimized configuration files, and extracts candidate configuration files when a preset baseline configuration file exists in the package.

[0018] Step S3 calculates the compatibility between candidate configuration files and target devices based on multi-dimensional profile information including device profile, user profile, and configuration file quality profile, determines the compatible target configuration file, and performs a pre-evaluation of optimization effect.

[0019] Step S4, based on the processing results of steps S2 to S3, configures the installation mode, generates a metadata file, and submits it for installation together with the application package;

[0020] Step S5 monitors the generation of runtime configuration files for installed applications, obtains the configuration files and attaches multi-dimensional profile information, compares them with existing configuration files in the cloud, and uploads them when the differences meet the conditions for subsequent installation and reuse.

[0021] In a possible preferred embodiment, the pre-evaluation of the optimization effect includes at least one of the following:

[0022] The expected performance gains are evaluated based on the overlap between the application package code structure and the hotspots of the candidate configuration files.

[0023] Based on historical optimization data of similar equipment, predict the benefits of this optimization;

[0024] Based on the configuration file size, number of hotspots, and device performance, predict compilation time and storage increment;

[0025] Based on the verification information of configuration files and application packages, as well as system status, the risk of compilation failure is predicted.

[0026] In a possible preferred embodiment, the configured installation mode includes:

[0027] Fast Installation Mode: Performs a normal installation when no optimization configuration file is available and no baseline configuration file is available;

[0028] Smart Install Mode: When no optimization configuration file is available but the application package contains a baseline configuration file, perform optimized installation based on the baseline configuration file;

[0029] Performance-based installation mode: When a matching optimization configuration file exists, an optimized installation is performed based on the optimization configuration file.

[0030] If the pre-evaluation of the optimization effect is lower than expected and the user is currently in the fast installation preference scenario, the system will revert to either the smart installation mode or the fast installation mode.

[0031] In a possible preferred embodiment, when the application enters an upgrade scenario, the steps further include:

[0032] Identify older versions of the same application package, older configuration files, and newer versions of the application package;

[0033] Compare the code structure of old and new versions of the application package to establish cross-version mapping relationships for hot code.

[0034] Migrate hot items with verifiable mappings from the old version configuration file to the new version candidate set; perform either elimination or de-weighting on hot items with unverifiable mappings.

[0035] The migrated hotspot set is merged with either the new version's pre-built baseline configuration file or an external optimized configuration file to form a new version candidate configuration file. The migration credibility and expected benefits are evaluated, and those that meet the criteria are given priority for upgrade installation.

[0036] In a possible preferred embodiment, the method further includes a scene-aware dynamic optimization step:

[0037] Obtain the current context state of the terminal, wherein the context includes at least one of the following: operating mode, device state, and user interaction state;

[0038] Filter the candidate configuration files to select the target configuration file, compilation filter, and delayed compilation strategy that best match the current context;

[0039] Depending on the context, execute any of the following processing steps: target configuration file, compiler filter, dynamic switching between delayed compilation strategies, downgrade or supplementary optimization.

[0040] In a possible preferred embodiment, the method further includes the step of initial installation preheating to generate a configuration file:

[0041] When installing for the first time and no external optimization configuration files are available, select any one of the following paths as representative paths: startup path, first screen rendering path, list rendering path, or details page jump path;

[0042] Run the representative usage path in the background using either a shadow instance or a controlled silent execution mode;

[0043] Collect hotspot information and critical startup paths during the execution process to generate initial candidate configuration files;

[0044] Depending on system resource conditions, you can choose to immediately trigger lightweight optimization compilation or reserve it for subsequent upgrade input;

[0045] Based on the incremental hotspot differences after actual use, the initial candidate configuration files are modified.

[0046] In a possible preferred embodiment, the method further includes a step of intelligent upgrade and optimization compilation:

[0047] Mark applications that are not currently being optimized and monitor their local configuration file generation process;

[0048] If the current version of the application already has a corresponding external optimization configuration file, then perform optimization compilation;

[0049] If the local configuration file has been generated, then optimize and compile it, and collect and upload the data.

[0050] If optimization compilation is to be performed, the candidate configuration file that includes the external optimization configuration file will be given priority, and a pre-evaluation of the optimization effect will be performed first.

[0051] If the system detects that it is currently in a preset condition window, it will perform an upgrade and optimization compilation; if the upgrade and optimization results in either failure to reach the preset threshold or an increased risk of failure, it will roll back to the previous stable strategy.

[0052] To achieve the above objectives, according to another aspect of the present invention, an Android application installation optimization system is also provided, comprising:

[0053] The application source access module is used to receive application packages to be installed from different application installation sources;

[0054] The first processing module is used to collect and upload runtime configuration files of applications installed on the terminal.

[0055] The second processing module is used to attach multi-dimensional profile information to the collected configuration files and to perform aggregation on configuration files of the same application and version.

[0056] The application installation package parsing module is used to parse the application's package information and extract the baseline configuration file;

[0057] The configuration file matching module is used to identify external configuration file resources corresponding to the target application, and, in conjunction with the baseline configuration file, to filter out the appropriate target configuration file based on multi-dimensional profile information.

[0058] The decision module is used to predict the expected benefits and costs of the target configuration file to select the installation mode, and encapsulate the target configuration file into a metadata file that can be consumed during the installation phase, and submit it for installation together with the application package;

[0059] The status tracking module is used to provide feedback on the application compilation and optimization process and results;

[0060] The upgrade and optimization module is used to generate and perform upgrade and optimization on unoptimized application monitoring configuration files.

[0061] The system scheduling module is used to control and optimize timing by combining equipment status and manufacturer strategies;

[0062] The version evolution module is used to quickly generate candidate configuration files for new versions based on old version configuration files when upgrading applications;

[0063] The scene awareness module is used to switch configuration files or compilation strategies based on the terminal context;

[0064] The preheating module is used to generate an initial candidate configuration file based on either a preset path or a shadow instance when the installation is first performed and there is a lack of available external configuration file resources.

[0065] To achieve the above objectives, according to another aspect of the present invention, a mobile terminal is also provided, comprising: a processor; a memory for storing one or more programs; and when the one or more programs are executed by the processor, causing the processor to perform the steps of the Android application installation optimization method as described above.

[0066] To achieve the above objectives, according to another aspect of the present invention, a computer-readable storage medium is also provided, on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the Android application installation optimization method as described above are implemented.

[0067] The Android application installation optimization method, system, mobile terminal, and storage medium provided by this invention address the technical shortcomings of existing installation systems, such as limited normal installation capabilities, insufficient profile optimization data management, inconsistent optimization behavior across multiple application sources, lack of pre-compilation benefit prediction, and the neglect of device and user differences in unified profiles for the same version. By uniformly completing profile collection, matching, evaluation, and optimization installation within the installation system, different application sources can share the same underlying profile assets and optimization installation capabilities. It employs an evaluation-before-compilation mechanism and combines multi-dimensional profile information to achieve joint adaptation and screening, achieving unified optimization across multiple sources and profile optimization. The specific technical effects of e-asset closed-loop accumulation, differentiated and precise matching, risk-controlled compilation decisions, and flexible installation modes can comprehensively improve the performance of application cold start, first screen display, and page switching, optimize the user installation experience, and have good compatibility without relying on Google Play or GMS. It can cover applications with low installation volume, private distribution, and non-store sources, and can also form a continuous optimization closed loop of collection, uploading, aggregation, and reuse. Compared with existing technologies, this solution converges optimization capabilities into the installation system, creates a unified optimization foundation shared across sources, and realizes a decoupled architecture in which the market only cares about distribution and the system is responsible for optimization. This is fundamentally different from traditional solutions that have decentralized optimization capabilities or rely on Google Cloud Profile. Attached Figure Description

[0068] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0069] Figure 1 : This is a schematic diagram of the system architecture for adapting to multiple application installation sources in the installation optimization system and method of the present invention;

[0070] Figure 2This is a schematic diagram illustrating the process of runtime profile collection and uploading in the installation optimization system and method of the present invention.

[0071] Figure 3 This is a schematic diagram illustrating the process of receiving APKs from multiple sources in the installation optimization system and method of the present invention.

[0072] Figure 4 This is a schematic diagram of the installation mode decision-making process in the installation optimization system and method of the present invention.

[0073] Figure 5 This is a schematic diagram of the DM file generation and installation session submission process in the installation optimization system and method of the present invention.

[0074] Figure 6 This is a schematic diagram of the intelligent upgrade optimization compilation process in the installation optimization system and method of the present invention;

[0075] Figure 7 This is a schematic diagram illustrating the joint matching process of device profile, user profile, and profile profile in the installation optimization system and method of the present invention.

[0076] Figure 8 This is a schematic diagram of the pre-evaluation process of the optimization effect in the installation optimization system and method of the present invention;

[0077] Figure 9 This is a schematic diagram of the ROM system collaborative scheduling and upgrade optimization compilation process in the installation optimization system and method of the present invention;

[0078] Figure 10 This is a schematic diagram illustrating the rapid migration process from an old version Profile to a new version candidate Profile in the installation optimization system and method of the present invention.

[0079] Figure 11 This is a schematic diagram of the scene-aware Profile switching process in the installation optimization system and method of the present invention.

[0080] Figure 12 This is a schematic diagram of the process of generating a profile during the initial installation of a shadow instance in the installation optimization system and method of the present invention.

[0081] Explanation of reference numerals in the attached figures

[0082] 100 represents the overall installation system; 110 represents the application source access module; 120 represents the runtime profile monitoring and collection module; 130 represents the profile construction and aggregation module; 140 represents the installation package parsing module; 150 represents the application configuration file matching module; 160 represents the Baseline Profile extraction module; 170 represents the optimization effect pre-evaluation module; 180 represents the installation mode decision module; 190 represents the DM file generation module; 200 represents the installation execution module; 210 represents the dexopt compilation optimization status tracking module; 220 represents the intelligent upgrade optimization compilation module; and 230 represents the ROM system. The modules are: Cooperative Scheduling Module, 240 (Cross-Version Profile Evolution Module), 250 (Scene-Aware Profile Switching Module), 260 (First-Installation Shadow Warm-up Module), 300 (Cloud or Backend Service), 310 (General Aggregated Profile), 320 (Device Group Aggregated Profile), 330 (User Group Aggregated Profile), 340 (Scene Profile), 350 (Candidate Profile Generated by Cross-Version Migration), 400 (Old Version APK), 410 (Old Version Profile), 420 (New Version APK), and 430 (New Version Candidate Profile). Detailed Implementation

[0083] To enable those skilled in the art to better understand the technical solutions of the present invention, the specific technical solutions of the present invention will be clearly and completely described below in conjunction with embodiments, so as to help those skilled in the art further understand the present invention. Obviously, the embodiments described in this application are merely some embodiments of the present invention, and not all embodiments. It should be noted that, for those skilled in the art, the embodiments and features in the embodiments of this application can be combined with each other without departing from the concept of the present invention and without conflict. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the disclosure and protection scope of the present invention.

[0084] Furthermore, the terms "first," "second," "S1," "S200," etc., used in the specification, claims, and drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such features can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those described herein. At the same time, the stages described in each step are not necessarily to be implemented in the same step; it should be understood that the implementation order of the contents of each step stage can be adjusted and interchanged without violating the inventive concept, so that embodiments of the invention described herein can be implemented in orders other than those described herein. Additionally, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. Unless otherwise expressly specified and limited, the terms "set," "arrange," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; a mechanical connection or an electrical connection; a direct connection or an indirect connection through an intermediate medium; or a connection within two elements. Those skilled in the art can understand the specific meaning of the above terms in this case based on the specific circumstances and in conjunction with existing technology.

[0085] Given that the existing Android installation system lacks a mechanism that can adapt to multiple app stores and APK sources, and uniformly complete the collection, uploading, and optimization of application runtime profiles during the installation phase, it is difficult to accumulate and reuse application profiles, and the installation behavior is difficult to standardize, resulting in unstable performance improvement after the first installation of the application.

[0086] Therefore, as Figures 1 to 12 As shown, this invention provides an Android application installation optimization system, examples of which include:

[0087] -Application Source Access Module

[0088] This application source access module is used to receive application packages to be installed from different application installation sources. This module is primarily used to adapt to different upper-layer entry points, including but not limited to: proprietary application markets; third-party application markets; enterprise application distribution platforms; download managers; local APK files in file managers; and other callers that can pass APK paths, package names, version numbers, or related installation parameters to the installer.

[0089] The core function of this module is to convert installation requests from different sources into a unified installation task model within the installer.

[0090] -First Processing Module

[0091] The first processing module, namely the installed application runtime profile monitoring and collection module, is used to identify applications installed on the terminal, monitor target applications, and perform runtime profile collection. The collected profiles are named according to a specific format and uploaded to the cloud server.

[0092] For example, in this example implementation, the installation system can monitor the target application's generation of runtime profiles through ArtManager or an equivalent runtime interface, and promptly obtain the corresponding profiles. The collected profiles are renamed according to the "package name-version number" rule; the profiles can be uploaded to the enterprise backend, object storage, cloud services, or other reusable carriers for subsequent installations; the profiles are compared with existing profiles of the same name in the "carrier," and if the difference exceeds a certain threshold (e.g., 20%, which is compared by converting them from binary to readable text format using Profman), the profile is uploaded to the "carrier" for further processing to avoid excessive invalid uploads.

[0093] At the same time, further device profile information corresponding to this profile is collected, including but not limited to SoC model, ABI, number of CPU cores, clock speed, memory capacity, flash memory type, Android version, screen resolution, and battery and temperature status.

[0094] Further collect user profile information corresponding to this profile, including but not limited to cold start frequency, session duration distribution, startup time distribution, page dwell time distribution, scrolling operation ratio, and notification activation ratio. However, it is preferred to upload the information in a de-identified or tagged form rather than uploading the original behavior details.

[0095] Further calculate the profile quality profile, including but not limited to startup link coverage, hotspot stability, method concentration, file size, source credibility, and the most recent effective benefit.

[0096] Through this first processing module, the installation system can not only consume external profiles, but also continuously accumulate and output new profile assets.

[0097] -Second Processing Module

[0098] The second processing module, namely the configuration file profile building and aggregation module, is used to attach multi-dimensional profile information to the collected configuration files and to perform aggregation on configuration files of the same application and version.

[0099] Specifically, this module is generally independent of the system, such as an enterprise backend, object storage, cloud services, or other carriers that can be reused in subsequent installations. It is responsible for aggregating uploaded configuration files (Profiles) for the same application and version, extracting more frequently used configurations, and preparing them for distribution at any time.

[0100] For example, in this example implementation, the newly uploaded configuration file (Profile) meets the aggregation requirements with the existing configuration file (Profile) with the same name, that is, the difference between the two exceeds a certain threshold (e.g., 20%). Then, Profman converts them from binary to readable text format for comparison, and extracts and records hotspot classes, hotspot methods, startup-related methods and page critical path markers to form a structured hotspot set.

[0101] Then, based on the device profile, the profile is divided into multiple device clusters, which can be constructed according to dimensions such as SoC, ABI, internal save location, Android version, and storage performance level; based on the user usage profile, the profile is divided into multiple behavioral clusters, which may include cold start priority, content browsing, direct message delivery, and low-frequency high-interaction, etc.; based on the profile quality profile, candidate profiles are sorted and filtered, and profiles with abnormally large values, low coverage, poor stability, or low historical performance are removed.

[0102] Finally, if any changes occur to the more frequent (e.g., above 10%) parts of the hotspot set, they are merged into a new text, which is then converted into a new profile using Profman to replace the old profile. Ideally, at least three types of aggregation results should be generated: a general aggregation profile, a device-specific aggregation profile, and a user behavior-specific aggregation profile, for selection based on priority during subsequent matching.

[0103] -Application installation package parsing module

[0104] The application installation package parsing module includes: an installation package parsing module and an application baseline profile extraction module.

[0105] The installation package parsing module is used to parse the application's package information, such as package name, version number, version name, APK file path, file size, and other metadata information used for subsequent installation decisions from the APK to be installed. Simultaneously, the installer uses this information to establish a unified identifier for the target APK, providing a basis for subsequent profile matching and pattern selection.

[0106] The application benchmark configuration file extraction module is used to detect and extract preset optimization data from the APK, preferably detecting at least one of the following paths:

[0107] 1.`assets / dexopt / baseline.prof`;

[0108] 2.`assets / dexopt / baseline.profm`;

[0109] 3.`assets / dexopt / baseline-prof.txt`.

[0110] If the corresponding file is detected, it will be extracted and used as input for installation optimization.

[0111] -Configuration file matching module

[0112] The configuration file matching module, also known as the application configuration file (Profile) matching module, is used to identify external configuration file resources corresponding to the target application. These external configuration file resources can come from:

[0113] 1. Application profile pre-generated by application developers or testing processes;

[0114] 2. Application configuration files (Profiles) distributed by the configuration file (Profile) aggregation module;

[0115] 3. The application configuration file (Profile) distributed by the app store along with the APK;

[0116] 4. Application profiles shared by enterprise platforms, cloud services, or other terminals;

[0117] 5. New version candidate profiles are quickly generated by the cross-version profile evolution module based on the old version profile.

[0118] This example prioritizes the installation system based on the "package name-version number" rule to identify and filter application profile resources, selects the profile that matches the target application version, and then selects the preferred profile according to the priority order mentioned above.

[0119] Furthermore, the application configuration file matching module preferably employs a multi-dimensional profile information suitability scoring mechanism to filter suitable target configuration files, which includes at least the following factors:

[0120] 1. Version compatibility;

[0121] 2. Apply signature or DEX verification to check the match.

[0122] 3. Device profile similarity;

[0123] 4. User profile similarity;

[0124] 5. Profile quality rating;

[0125] 6. Historical return score;

[0126] 7. Compilation time prediction penalty.

[0127] The installation system can calculate the total fit score of candidate profiles based on the above factors, and select the profile with the highest score that meets the threshold to enter the subsequent pre-evaluation process.

[0128] - Decision Module

[0129] The decision module is used to predict the expected benefits and costs of the target configuration file to select the installation mode, and encapsulate the target configuration file into a metadata file that can be consumed during the installation phase, and submit it for installation together with the application package.

[0130] Specifically, the decision-making module example includes: optimization effect pre-evaluation module, installation mode decision module, DM file generation module, and installation execution module.

[0131] The optimization effect pre-evaluation module is used to predict the adaptation effect of candidate profiles before the DM file is officially generated and dexopt is triggered. Preferably, the prediction includes at least one or more of the following:

[0132] 1. Calculate the expected revenue score based on the overlap between the DEX structure in the APK, the candidate profile hotspot set, and the critical startup path;

[0133] 2. Based on historical startup time, first screen time, stuttering frame rate, or post-upgrade revenue data for the same device and user groups, predict the benefits of this optimization.

[0134] 3. Based on candidate profile size, number of hot methods, and device CPU and storage performance, predict compilation time and storage increment;

[0135] 4. Based on the verification information of the Profile and the target APK, historical failure records, and the current system status, predict the risk of compilation failure;

[0136] 5. Perform "shadow evaluation" on multiple candidate profiles, that is, perform lightweight simulation scoring based on textual hotspot sets and historical compilation results without formally submitting the installation.

[0137] 6. Perform a lightweight benchmark evaluation on the candidate profile based on a pre-defined set of representative use cases. The set of representative use cases may include at least cold start scenario, first screen rendering scenario, list scrolling scenario, and detail page jump scenario.

[0138] In a preferred embodiment, if the revenue score is lower than the threshold, or the compilation time prediction is higher than the threshold and the user is currently in a fast installation preference scenario, the system reverts to the smart installation mode or the fast installation mode.

[0139] The installation mode decision module is used to select an installation strategy based on available optimization data. Preferably, it includes the following three modes:

[0140] Fast Installation Mode: Perform a normal installation when the installation system does not detect an available external profile and the APK does not contain a Baseline Profile.

[0141] Smart Installation Mode: When the installation system does not detect a matching external profile, but detects a Baseline Profile included in the APK, it extracts the Baseline Profile and performs optimized installation.

[0142] Performance Installation Mode: When the installation system detects an external profile that matches the target APK package name and version number, it will prioritize using that external profile for optimized installation; if there are multiple candidate profiles, the target profile will be selected by combining the profile matching degree and the pre-evaluation results.

[0143] Specifically, the DM file generation module, in order to enable the system to directly consume the optimized data during the installation phase, encapsulates the selected Profile into a DM file. For example, the DM file preferably uses a ZIP structure; the ZIP contains a `primary.prof` file; ZIP entries are stored uncompressed; size information and CRC checksums are pre-written; the DM file corresponds to `base.apk`, for example, named `base.dm`.

[0144] The installation execution module submits the APK and DM file together via the system installation session interface. For example, it preferably writes the APK to `base.apk` in the installation session and the DM file to `base.dm` in the installation session; then it submits an installation request, which the system uses to perform compilation optimization during the installation and subsequent dexopt phases.

[0145] -State Tracking Module

[0146] The state tracking module, also known as the dexopt compilation optimization state tracking module, is used to provide phased feedback on the dexopt compilation optimization process, including: the dexopt optimization phase and the optimization success or failure phase.

[0147] - Upgrade and optimization module

[0148] The upgrade and optimization module, also known as the intelligent upgrade and optimization compilation module, is used to monitor and match configuration files for applications in the fast installation mode, and promptly remind users to upgrade to the performance installation mode.

[0149] -System Scheduling Module

[0150] The system scheduling module, also known as the ROM system collaborative scheduling module, is used to schedule and control the optimization compilation process by leveraging the system-level privileges and vendor capabilities of the installed system. Preferably, this module includes at least the following capabilities:

[0151] 1. Obtain device charging status, power threshold, temperature range, free storage space, and system free window;

[0152] 2. Obtain system-side cold start statistics, application recent startup time, compilation history, and failure history;

[0153] 3. On system versions that support ART Service, adjust the compilation filter, background dexopt scheduling timing, or specified application dexopt triggering strategy for specific applications through system properties, system APIs, or callback mechanisms;

[0154] 4. Select the appropriate processing method based on the manufacturer's ROM policy: "Optimize immediately", "Optimize when idle", "Only remind user to optimize", or "Cancel optimization and roll back".

[0155] 5. Perform downgrades, circuit breakers, or rollbacks on low-yield or high-risk candidate profiles.

[0156] - Version Evolution Module

[0157] The version evolution module, also known as the cross-version profile evolution module, is used to quickly generate a candidate profile for the new version based on optimized assets when an application upgrades from an older version to a new version, without waiting for the new version to regenerate the runtime profile. Preferably, this module includes at least the following processing steps:

[0158] 1. Identify older versions of the same application's APK, older version profile, and newer version APK;

[0159] 2. Compare the DEX structures of the old and new versions of the APK, and build cross-version mapping relationships between classes, methods, DEX files, and hot paths;

[0160] 3. Migrate the hot classes and methods in the old version of Profile, and retain the hot items that still exist in the new version and whose similarity meets the threshold;

[0161] 4. Remove or demote hot items that have been deleted, renamed, have undergone significant inline changes, or whose mapping relationships cannot be confirmed;

[0162] 5. Merge the migrated hotspot set with the BaselineProfile, cloud-side aggregated Profile, or scenario Profile that come with the new version APK to form a new version candidate Profile;

[0163] 6. Calculate the migration confidence score for the candidate profiles of the new version, and give priority to using them for upgrade installation or the first optimization compilation after upgrade if the score is higher than the threshold;

[0164] In a preferred embodiment, the construction of the cross-version mapping relationship is based on at least one or more of the following:

[0165] 1. An exact match between the class name, method signature, and the location of the method in the DEX file;

[0166] 2. Similarity matching based on class hierarchy, package path, call relationship, or control flow summary;

[0167] 3. Method body hash, basic block hash, instruction sequence normalized hash, or incremental patch information;

[0168] 4. Differential packages, patch packages, or symbol mapping information exported during the build phase between old and new versions.

[0169] - Scene awareness module,

[0170] The scene-aware module, also known as the scene-aware profile switching module, is used to select the most suitable profile from multiple candidate profiles or multiple compilation strategies based on the current context of the terminal. Preferably, the context includes at least one or more of the following:

[0171] 1. Game mode, high refresh rate mode, power saving mode, car mode, screen mirroring mode;

[0172] 2. Foreground application type, network environment, battery status, and temperature status;

[0173] 3. User interaction rhythm, notification activation mode, and session duration distribution.

[0174] In a preferred embodiment, the installation system can select different aggregated profiles, different candidate hotspot sets, or different compilation filters in different scenarios to achieve scenario-based optimization.

[0175] -Preheating module

[0176] The preheating module, also known as the first-installation shadow preheating module, is used to generate an initial candidate configuration file based on a preset path or shadow instance when the installation is first performed and there is a lack of available external configuration file resources.

[0177] For example: Before the user actually clicks the application icon, the system silently executes representative startup paths, first screen rendering paths, list rendering paths, or details page jump paths in the background; based on the above silent execution process, an initial Runtime Profile or startup hotspot set is generated; if the system resources meet the conditions, a lightweight optimization compilation is directly triggered using the initial Runtime Profile; if the system resources are insufficient, the initial Runtime Profile is used as the input for subsequent upgrade optimization compilation.

[0178] For the above optimized system example, refer to Figures 1 to 12 As shown, the present invention also provides an Android application installation optimization method, the steps of which include:

[0179] Step S1 receives the application package to be installed from application installation sources such as application markets, enterprise distribution platforms or local file entry points, and parses the identification information of the application package, such as package name, version number and file information.

[0180] Step S2 retrieves external configuration file resources (including external profiles, aggregated profiles, or profiles generated by cross-version migration) based on the identification information, filters out optimized configuration files, and extracts them when a preset baseline configuration file exists in the package, so that the optimized configuration file and the baseline configuration file are used as candidate configuration files.

[0181] Step S3 calculates the compatibility between candidate configuration files and target devices based on multi-dimensional profile information including device profile, user profile, and configuration file quality profile. For candidate configuration files that meet the compatibility threshold, a pre-evaluation of the optimization effect is performed, including optimization benefits, compilation time, storage increment, and failure risk.

[0182] In this example, the optimization effect pre-evaluation instance includes at least one of the following:

[0183] The expected performance gains are evaluated based on the overlap between the application package code structure and the hotspots of the candidate configuration files.

[0184] Based on historical optimization data of similar equipment, predict the benefits of this optimization;

[0185] Based on the configuration file size, number of hotspots, and device performance, predict compilation time and storage increment;

[0186] Based on the verification information of configuration files and application packages, as well as system status, the risk of compilation failure is predicted.

[0187] Step S4 determines the installation mode based on the processing results of steps S2 to S3, and generates a metadata file and submits it for installation together with the application package.

[0188] Specifically, in a preferred embodiment, the installation mode includes:

[0189] Fast Installation Mode: Performs a normal installation when no optimization configuration file is available and no baseline configuration file is available;

[0190] Smart Install Mode: When no optimization configuration file is available but the application package contains a baseline configuration file, perform optimized installation based on the baseline configuration file;

[0191] Performance-based installation mode: When a matching optimization configuration file exists, an optimized installation is performed based on the optimization configuration file.

[0192] If the pre-evaluation of the optimization effect is lower than expected and the user is currently in the fast installation preference scenario, the system will revert to either the smart installation mode or the fast installation mode.

[0193] Users can choose a default installation mode according to their preferences, such as the system defaulting to "Performance Installation". When an application does not meet the requirements of a certain installation mode, the application's installation mode will fall back from the current default mode according to the priority of "Performance Installation" -> "Smart Installation" -> "Fast Installation" until the conditions are met.

[0194] Step S5 monitors the generation of runtime configuration files for installed applications, obtains the configuration files and attaches multi-dimensional profile information, compares them with existing configuration files in the cloud, and uploads them when the differences meet the conditions for subsequent installation and reuse.

[0195] Specifically, step S5, which involves the collection and uploading of runtime profiles, includes the following examples:

[0196] Step S501: Install the system monitoring system to generate runtime profiles for applications already installed in the system;

[0197] Step S502: When an application generates a runtime profile, the system directly obtains the profile using its higher privileges.

[0198] Step S503: Rename the collected Profile according to the application package name and version number in a specific format, and attach device profile, user profile and Profile quality profile tags.

[0199] Step S504: If there is currently no configuration file (Profile) with the same name in the enterprise backend, cloud service or other reusable carrier, then upload it directly to the enterprise backend, cloud service or other reusable carrier.

[0200] Step S505: Otherwise, both profiles are converted from binary to text format using Profman for comparison. If the difference exceeds a certain threshold (e.g., 20%), the collected profile is uploaded to the enterprise backend, cloud service, or other reusable carrier.

[0201] The steps for reusing a uploaded Profile include:

[0202] Step S506: Collect and upload the required configuration file from the Android device and store it in a searchable carrier;

[0203] Step S507: Compare the newly uploaded configuration file with the configuration file of the same name in the "carrier" by converting it into text format using Profman, and extract and record the hotspot classes, hotspot methods, startup-related methods and page critical path markers to form a structured hotspot set.

[0204] Step S508: The structured hotspot set is grouped according to the device profile, user profile, and profile quality profile;

[0205] Step S509: If there are any changes to the parts of the hotspot set that have a high frequency (e.g., higher than 10%), merge them into a new text, and then use Profman to convert this text into a new aggregation configuration file to replace the old aggregation configuration file.

[0206] In step S510, when a subsequent installation task occurs, the installation system reads the aggregate configuration file corresponding to the APK to be installed from the carrier;

[0207] Step S511: The installation system filters the target profile based on the current target device profile, user profile, and historical revenue information;

[0208] Step S512, if necessary, perform lightweight baseline prediction on the target profile based on a pre-set representative scene set;

[0209] Step S513: Use the target profile for performance installation or participate in the installation decision together with other optimization data.

[0210] Furthermore, in an optional embodiment, the method of the present invention further includes: an intelligent upgrade and optimization compilation step:

[0211] Step S601: The installation system marks applications that are not currently optimized and compiles, and monitors the local configuration file generation process for them;

[0212] Step S602: If the current version of the application already has a corresponding cloud-side configuration file or a candidate profile generated by cross-version migration, then the user can be notified to perform optimization compilation.

[0213] Step S603: If it is detected that their local configuration files have been generated, notify the user to perform optimization compilation and prepare to collect and upload data according to the aforementioned requirements;

[0214] Step S604: If the user decides to perform optimized compilation, the cloud-side configuration file or the candidate profile generated by cross-version migration will be given priority, and a pre-evaluation will be performed first.

[0215] Step S605: If the system detects that it is currently in a charging, idle, low temperature and sufficient storage window, it can automatically perform upgrade optimization compilation;

[0216] Step S606: If the effect of the upgrade and optimization does not reach the preset threshold or the risk of failure increases, roll back to the previous stable strategy.

[0217] Furthermore, in an optional embodiment, the method of the present invention further includes the step of quickly generating a new version profile when the application enters an upgrade scenario:

[0218] Step S701: The installation system identifies the old version APK, old version Profile, and new version APK of the application to be updated;

[0219] Step S702: parse the DEX structure, class path, method signature, and hotspot entry path of the old version APK and the new version APK;

[0220] Step S703: Based on complete matching, similarity matching, differential packet information, or constructed mapping information, establish a mapping relationship from old version hot items to new version hot items;

[0221] Step S704: Migrate the hotspot classes and hotspot methods that can be confirmed to be mapped in the old version Profile to the candidate hotspot set in the new version, and remove, downgrade or mark the hotspot items that cannot be confirmed to be mapped.

[0222] Step S705: The new version candidate hotspot set is merged with the new version APK's built-in Baseline Profile, cloud-side aggregated Profile, or scenario Profile to form a new version candidate Profile;

[0223] Step S706: Calculate the migration confidence, expected benefits, and failure risk for the new version candidate profile;

[0224] Step S707: If the migration confidence and expected benefits reach the threshold, then during the upgrade installation, the new version candidate Profile will be used first to perform optimized installation or the first optimized compilation after the upgrade.

[0225] Step S708: After the new version is actually running, collect incremental Runtime Profiles and perform corrections, additions, or replacements on the candidate Profiles for the new version.

[0226] Furthermore, in an optional embodiment, the method of the present invention further includes a scene-aware dynamic optimization step:

[0227] Step S801: The system installs the current context state of the terminal, including the operating mode, device status, and user interaction status.

[0228] Step S802: Select the target profile that best matches the current context from the general aggregated profile, device cluster aggregated profile, user cluster aggregated profile, scenario profile or candidate profile generated by cross-version migration;

[0229] Step S803: Select the target profile, compilation filter, or delayed compilation strategy based on the current context;

[0230] Step S804: When the context changes, perform switching, downgrading, or supplementary optimization on the target profile or compilation strategy.

[0231] Furthermore, in an optional embodiment, the method of the present invention further includes the step of initial installation and preheating to generate a configuration file:

[0232] Step S901: When the application is installed for the first time and there is no available external profile, the installation system selection includes any one of the following: startup path, first screen rendering path, list rendering path, or details page jump path, which is used as the representative path;

[0233] Step S902: Run the representative usage path in the background as a shadow instance, virtual execution instance, or controlled silent execution mode;

[0234] Step S903: Collect hot classes, hot methods and critical startup paths during the silent execution process to generate an initial candidate profile;

[0235] Step S904: If the current system state meets the conditions, then use the initial candidate Profile to trigger lightweight optimization compilation;

[0236] Step S905: If the difference between the incremental hotspots collected after a real user starts the program and the initial candidate profile exceeds a threshold, then the initial candidate profile is corrected and re-scored.

[0237] Corresponding to the above examples, the present invention also provides a mobile terminal, comprising: a processor; a memory for storing one or more programs; and when the one or more programs are executed by the processor, causing the processor to implement the steps of the Android application installation optimization method as described in any of the above examples.

[0238] Corresponding to the above examples, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the steps of the Android application installation optimization method as described above.

[0239] In summary, the Android application installation optimization method and system, mobile terminal, and storage medium provided by this invention address the technical shortcomings of existing installation systems, such as limited normal installation capabilities, insufficient profile optimization data management, inconsistent optimization behavior across multiple application sources, lack of pre-compilation benefit prediction, and the neglect of device and user differences in unified profiles for the same version. By uniformly completing profile collection, matching, evaluation, and optimization installation within the installation system, different application sources share the same underlying profile assets and optimization installation capabilities. It employs an evaluation-before-compilation mechanism and combines multi-dimensional profile information to achieve joint adaptation and screening, thus achieving unified optimization across multiple sources and improved performance. The specific technical effects of closed-loop file asset accumulation, differentiated and precise matching, risk-controlled compilation decisions, and flexible installation modes can comprehensively improve the performance of application cold start, first screen display, and page switching, optimize the user installation experience, and have good compatibility without relying on Google Play or GMS. It can cover applications with low installation volume, private distribution, and non-store sources, and can also form a continuous optimization closed loop of collection, uploading, aggregation, and reuse. Compared with existing technologies, this solution converges optimization capabilities into the installation system, creates a unified optimization foundation shared across sources, and realizes a decoupled architecture in which the market only cares about distribution and the system is responsible for optimization. This is fundamentally different from traditional solutions that have decentralized optimization capabilities or rely on Google Cloud Profile.

[0240] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The present invention is limited only by the claims and their full scope and equivalents. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

[0241] Those skilled in the art will understand that, besides implementing the system, apparatus, unit, and its modules provided by this invention in purely computer-readable program code, the same program can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system, apparatus, and its modules provided by this invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; alternatively, modules for implementing various functions can be considered both software programs implementing the method and structures within the hardware component.

[0242] Furthermore, all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a microcontroller, chip, or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0243] Furthermore, various different implementations of the present invention can be combined arbitrarily, as long as they do not violate the spirit of the present invention, they should also be regarded as the content disclosed in the present invention.

Claims

1. An Android application installation optimization method, comprising the following steps: Step S1 receives application packages to be installed from different application installation sources and parses the identification information of the application packages; Step S2 retrieves external configuration file resources based on the identification information, filters out optimized configuration files, and extracts candidate configuration files when a preset baseline configuration file exists in the package. Step S3 calculates the compatibility between candidate configuration files and target devices based on multi-dimensional profile information including device profile, user profile, and configuration file quality profile, determines the compatible target configuration file, and performs a pre-evaluation of optimization effect. Step S4, based on the processing results of steps S2 to S3, configures the installation mode, generates a metadata file, and submits it for installation together with the application package; Step S5 monitors the generation of runtime configuration files for installed applications, obtains the configuration files and attaches multi-dimensional profile information, compares them with existing configuration files in the cloud, and uploads them when the differences meet the conditions for subsequent installation and reuse.

2. The method according to claim 1, wherein the pre-evaluation of the optimization effect includes at least one of the following: The expected performance gains are evaluated based on the overlap between the application package code structure and the hotspots of the candidate configuration files. Based on historical optimization data of similar equipment, predict the benefits of this optimization; Based on the configuration file size, number of hotspots, and device performance, predict compilation time and storage increment; Based on the verification information of configuration files and application packages, as well as system status, the risk of compilation failure is predicted.

3. The method according to claim 1, wherein the configured installation mode includes: Fast Installation Mode: Performs a normal installation when no optimization configuration file is available and no baseline configuration file is available; Smart Install Mode: When no optimization configuration file is available but the application package contains a baseline configuration file, perform optimized installation based on the baseline configuration file; Performance-based installation mode: When a matching optimization configuration file exists, an optimized installation is performed based on the optimization configuration file; If the pre-evaluation of the optimization effect is lower than expected and the user is currently in the fast installation preference scenario, the system will revert to either the smart installation mode or the fast installation mode.

4. The method according to claim 1, wherein when the application enters an upgrade scenario, the step further includes: Identify older versions of the same application package, older configuration files, and newer versions of the application package; Compare the code structure of old and new versions of the application package to establish cross-version mapping relationships for hot code. Migrate hot items with verifiable mappings from the old version configuration file to the new version candidate set; perform either elimination or de-weighting on hot items with unverifiable mappings. The migrated hotspot set is merged with either the new version's pre-built baseline configuration file or an external optimized configuration file to form a new version candidate configuration file. The migration credibility and expected benefits are evaluated, and those that meet the criteria are given priority for upgrade installation.

5. The method according to claim 1, further comprising a scene-aware dynamic optimization step: Obtain the current context state of the terminal, wherein the context includes at least one of the following: operating mode, device state, and user interaction state; Filter the candidate configuration files to select the target configuration file, compilation filter, and delayed compilation strategy that best match the current context; Depending on the context, execute any of the following processing steps: dynamic switching between target configuration files, compiler filters, delayed compilation strategies, downgrade, or supplementary optimization.

6. The method of claim 1, further comprising the step of generating a configuration file during initial installation preheating: When installing for the first time and no external optimization configuration files are available, select the following: Any one of the following paths can be used as a representative path: startup path, first screen rendering path, list rendering path, or details page navigation path. Run the representative usage path in the background either as a shadow instance or in a controlled silent execution mode; Collect hotspot information and critical startup paths during the execution process to generate initial candidate configuration files; Depending on system resource conditions, you can choose to immediately trigger lightweight optimization compilation or reserve it for subsequent upgrade input; Based on the incremental hotspot differences after actual use, the initial candidate configuration files are modified.

7. The method according to claim 1, further comprising the step of intelligent upgrade and optimization compilation: Mark applications that are not currently being optimized and monitor their local configuration file generation process; If the current version of the application already has a corresponding external optimization configuration file, then perform optimization compilation; If the local configuration file has been generated, then optimize and compile it, and collect and upload the data. If optimization compilation is to be performed, the candidate configuration file that includes the external optimization configuration file will be given priority, and a pre-evaluation of the optimization effect will be performed first. If the system detects that it is currently in a preset condition window, it will perform an upgrade and optimization compilation. If the upgraded and optimized results fail to meet the preset threshold or the risk of failure increases, the system will roll back to the previous stable strategy.

8. An Android application installation optimization system, comprising: The application source access module is used to receive application packages to be installed from different application installation sources; The first processing module is used to collect and upload runtime configuration files of applications installed on the terminal. The second processing module is used to attach multi-dimensional profile information to the collected configuration files and to perform aggregation on configuration files of the same application and version. The application installation package parsing module is used to parse the application's package information and extract the baseline configuration file; The configuration file matching module is used to identify external configuration file resources corresponding to the target application, and, in conjunction with the baseline configuration file, to filter out the appropriate target configuration file based on multi-dimensional profile information. The decision module is used to predict the expected benefits and costs of the target configuration file to select the installation mode, and encapsulate the target configuration file into a metadata file that can be consumed during the installation phase, and submit it for installation together with the application package; The status tracking module is used to provide feedback on the application compilation and optimization process and results; The upgrade and optimization module is used to generate and perform upgrade and optimization on unoptimized application monitoring configuration files. The system scheduling module is used to control and optimize timing by combining equipment status and manufacturer strategies; The version evolution module is used to quickly generate candidate configuration files for new versions based on old version configuration files when upgrading applications; The scene awareness module is used to switch configuration files or compilation strategies based on the terminal context; The preheating module is used to generate an initial candidate configuration file based on either a preset path or a shadow instance when the installation is first performed and there is a lack of available external configuration file resources.

9. A mobile terminal, comprising: processor; Memory, used to store one or more programs; When the processor executes the one or more programs, the processor performs the steps of the Android application installation optimization method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the Android application installation optimization method as described in any one of claims 1 to 7.

Images