Cloud phone bottom layer performance optimization method, device and equipment and storage medium

Abstract

This invention belongs to the field of cloud phone technology and discloses a method, apparatus, device, and storage medium for optimizing the underlying performance of cloud phones. It includes detecting the current scenario to be optimized; if it is multi-system underlying adaptation optimization, it collects terminal system characteristic information, dynamically generates a configuration file containing end-cloud collaboration SDK interaction parameters and interface layout rules, and adjusts the SDK transmission protocol, data parsing logic, and interface interaction logic accordingly to achieve multi-system underlying compatibility; if it is high-concurrency scenario underlying performance optimization, it collects underlying hardware resource status data in real time, identifies business scenarios and priorities, dynamically allocates underlying hardware resources based on resource status and priority, and dynamically adjusts encoding / decoding parameters and transmission protocols in audio and video streams, performing data compression and redundancy correction under weak network conditions. The above methods improve multi-system compatibility, high-concurrency resource utilization efficiency, and end-cloud collaborative transmission performance through scenario-adaptive optimization strategies.

Description

Technical Field

[0001] This invention relates to the field of cloud phone technology, and in particular to a method, apparatus, device, and storage medium for optimizing the underlying performance of cloud phones. Background Technology

[0002] Cloud phones utilize virtualization technology to run the mobile operating system on cloud servers, providing users with a complete mobile phone experience. While existing optimization solutions address resource scheduling and data transmission, they still suffer from the following shortcomings: First, they struggle to simultaneously adapt to the underlying characteristics of multiple systems such as iOS, HarmonyOS, and Android, leading to sluggish interface interactions and poor functional compatibility. Second, in high-concurrency streaming and multi-application parallel scenarios, resource scheduling is not deeply integrated with the real-time status of the underlying hardware, easily resulting in resource allocation imbalances and container startup delays. Third, traditional encoding / decoding and transmission protocols in edge-cloud collaborative transmission are not optimized for cloud phones, resulting in high latency and frequent stuttering in weak network environments.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a method, apparatus, device, and storage medium for optimizing the underlying performance of cloud phones, aiming to solve the technical problems of poor multi-system compatibility, low resource scheduling efficiency in high-concurrency scenarios, and high latency in end-to-cloud collaborative transmission in the prior art.

[0005] To achieve the above objectives, the present invention provides a method for optimizing the underlying performance of cloud phones, the method comprising the following steps: Detect the current scene that needs optimization; When the current scenario to be optimized is multi-system underlying adaptation optimization, system characteristic information is collected when the terminal accesses the system. The system characteristic information includes operating system type, version, and underlying architecture characteristics. Based on the system characteristic information, an adaptive configuration file is dynamically generated. The configuration file includes the interaction parameters, interface layout rules and touch response thresholds of the end-to-cloud collaboration SDK. Adjust the transmission protocol and data parsing logic of the end-to-cloud collaboration SDK according to the configuration file, and calibrate the interface interaction logic to achieve multi-system underlying compatibility; When the current scenario to be optimized is the underlying performance optimization of a high-concurrency scenario, the status data of the underlying hardware resources of the cloud phone is collected in real time. The status data includes the utilization rate of CPU, GPU, memory and storage resources. Identify the currently running business scenario and its priority, including cloud gaming, office applications, or high-concurrency streaming; Dynamically allocate underlying hardware resources based on the resource status data and the priority of the business scenario; During audio and video streaming, the encoding and decoding parameters and transmission protocol are dynamically adjusted, and data compression and redundancy error correction operations are performed in weak network environments.

[0006] In one embodiment, adjusting the transmission protocol and data parsing logic of the cloud-edge collaboration SDK according to the configuration file, and calibrating the interface interaction logic, includes: The SDK's audio and command stream transmission adaptation parameters are dynamically adjusted according to the configuration file, as well as the layout of interface elements and touch response latency, to adapt to different screen forms.

[0007] In one embodiment, after the steps of adjusting the transmission protocol and data parsing logic of the cloud-edge collaboration SDK according to the configuration file and calibrating the interface interaction logic, the method further includes: It automatically detects the application's running status, interface response speed, and functional compatibility. If any abnormalities are found, it returns to regenerate the configuration file until the adaptation verification is passed.

[0008] In one embodiment, the dynamic allocation of underlying hardware resources based on the resource status data and the service scenario priority includes: Determine the resource allocation scheme based on the priority of business scenarios, allocate GPU resources to high-priority businesses, adjust CPU scheduling weights, and optimize memory caching strategies; The resource isolation module of the virtualization layer isolates and allocates underlying resources according to the resource allocation scheme.

[0009] In one embodiment, the method further includes: Continuously monitor changes in resource status. If resource utilization exceeds a preset threshold, re-trigger resource allocation decisions and dynamically adjust the resource allocation scheme.

[0010] In one embodiment, the dynamic adjustment of encoding / decoding parameters and transmission protocol includes: Employing H.265 or AV1 encoding / decoding technologies, combined with adaptive bitrate adjustment, reduces audio and video streaming latency; We have customized the data transmission logic based on the WebRTC protocol to optimize the real-time transmission performance of command streams and audio streams.

[0011] In one embodiment, the method further includes: When a single node's resources are insufficient, multiple core nodes and sub-nodes are linked to perform cross-node resource scheduling.

[0012] Furthermore, to achieve the above objectives, this invention also proposes a cloud phone underlying performance optimization system, which is applied to the cloud phone underlying performance optimization method described above. The system includes: The scene detection module is used to detect the current scene to be optimized. The multi-system adaptation and optimization module is used to collect system characteristic information when the terminal accesses the system when the current optimization scenario is multi-system underlying adaptation and optimization. The system characteristic information includes operating system type, version and underlying architecture characteristics. The multi-system adaptation and optimization module is used to dynamically generate an adaptation configuration file based on the system characteristic information. The configuration file includes the interaction parameters, interface layout rules, and touch response thresholds of the end-to-cloud collaboration SDK. The multi-system adaptation and optimization module is used to adjust the transmission protocol and data parsing logic of the end-to-cloud collaboration SDK according to the configuration file, and to calibrate the interface interaction logic to achieve multi-system underlying compatibility. The high-concurrency performance optimization module is used to collect the status data of the underlying hardware resources of the cloud phone in real time when the current scenario to be optimized is the underlying performance optimization of a high-concurrency scenario. The status data includes the utilization rate of CPU, GPU, memory and storage resources. The high-concurrency performance optimization module is used to identify the currently running business scenario and the priority of the business scenario, which includes cloud gaming, office applications, or high-concurrency streaming. The high-concurrency performance optimization module is used to dynamically allocate underlying hardware resources based on the resource status data and the priority of the business scenario; The high-concurrency performance optimization module is used to dynamically adjust the encoding and decoding parameters and transmission protocol during audio and video streaming, and to perform data compression and redundancy error correction operations in weak network environments.

[0013] Furthermore, to achieve the above objectives, the present invention also proposes a cloud phone underlying performance optimization device, which includes: a memory, a processor, and a cloud phone underlying performance optimization program stored in the memory and executable on the processor. The cloud phone underlying performance optimization program is configured to implement the steps of the cloud phone underlying performance optimization method described above.

[0014] In addition, to achieve the above objectives, the present invention also proposes a storage medium storing a cloud phone underlying performance optimization program, wherein when the cloud phone underlying performance optimization program is executed by a processor, it implements the steps of the cloud phone underlying performance optimization method described above.

[0015] This invention detects the current scenario to be optimized. If it's a multi-system underlying adaptation optimization, it collects terminal system characteristic information, dynamically generates a configuration file containing end-to-cloud collaboration SDK interaction parameters and interface layout rules, and adjusts the SDK transmission protocol, data parsing logic, and interface interaction logic accordingly to achieve multi-system underlying compatibility. If it's a high-concurrency scenario underlying performance optimization, it collects underlying hardware resource status data in real time, identifies business scenarios and priorities, dynamically allocates underlying hardware resources based on resource status and priorities, and dynamically adjusts encoding / decoding parameters and transmission protocols in audio and video streams, performing data compression and redundancy correction under weak network conditions. This approach, through scenario-adaptive optimization strategies, improves multi-system compatibility, high-concurrency resource utilization efficiency, and end-to-cloud collaborative transmission performance. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the first embodiment of the cloud phone underlying performance optimization method of the present invention; Figure 2 This is a schematic diagram of the multi-system underlying adaptation and optimization process in the cloud phone underlying performance optimization method of the present invention; Figure 3 This is a schematic diagram of the underlying performance optimization process in the cloud phone underlying performance optimization method of the present invention for high-concurrency scenarios; Figure 4 This is a structural block diagram of the first embodiment of the cloud phone underlying performance optimization system of the present invention.

[0017] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0019] This invention provides a method for optimizing the underlying performance of cloud phones, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of a cloud phone underlying performance optimization method according to the present invention.

[0020] In this embodiment, the underlying performance optimization method for cloud phones includes the following steps: Step S10: Detect the current scene to be optimized.

[0021] In this embodiment, the execution entity is a cloud phone underlying performance optimization device, which has functions such as data processing, data communication, and program execution. The cloud phone underlying performance optimization device can be a computer terminal device or other network device, or other devices with similar functions. This embodiment does not limit the scope of the implementation.

[0022] It should be noted that cloud phones use virtualization technology to run the phone's operating system on a cloud server, providing users with a complete mobile phone experience. While existing optimization solutions involve resource scheduling and data transmission, they still have the following shortcomings: First, they struggle to simultaneously adapt to the underlying characteristics of multiple systems such as iOS, HarmonyOS, and Android, leading to sluggish interface interactions and poor functional compatibility; second, in high-concurrency streaming and multi-application parallel scenarios, resource scheduling is not deeply integrated with the real-time status of the underlying hardware, easily resulting in resource allocation imbalances and container startup delays; third, traditional encoding / decoding and transmission protocols in edge-cloud collaborative transmission are not optimized for cloud phones, resulting in high latency and frequent stuttering in weak network environments.

[0023] To address the aforementioned technical issues, this embodiment detects the current scenario to be optimized. If it involves multi-system underlying adaptation optimization, terminal system characteristic information is collected, and a configuration file containing end-to-cloud collaboration SDK interaction parameters and interface layout rules is dynamically generated. Based on this, the SDK transmission protocol, data parsing logic, and interface interaction logic are adjusted to achieve multi-system underlying compatibility. If it involves high-concurrency scenario underlying performance optimization, underlying hardware resource status data is collected in real time, business scenarios and priorities are identified, and underlying hardware resources are dynamically allocated based on resource status and priority. Furthermore, encoding / decoding parameters and transmission protocols are dynamically adjusted in audio and video streams, and data compression and redundancy correction are performed under weak network conditions. This approach, through scenario-adaptive optimization strategies, improves multi-system compatibility, high-concurrency resource utilization efficiency, and end-to-cloud collaborative transmission performance. Specifically, it can be implemented as follows.

[0024] In its implementation, this embodiment constructs a four-layer architecture: a multi-system adaptation layer, a virtualization optimization layer, an edge-cloud collaboration layer, and a resource scheduling layer. This architecture achieves the underlying performance optimization goals of multi-system compatibility, efficient resource utilization, and low-latency edge-cloud transmission. The specific architecture includes: Multi-system adaptation layer: As the foundational layer for underlying performance optimization, its core components include a system feature analysis module, an edge-cloud collaboration software development kit (SDK) optimization module, and a user interface adaptation module. System feature analysis module: This module analyzes the underlying architectural features of systems such as iOS, HarmonyOS, and Android in real time (e.g., iOS's sandbox mechanism and HarmonyOS's distributed soft bus features) and generates system adaptation configuration files. Edge-cloud collaboration SDK optimization module: This module customizes SDK interaction logic for different system features, optimizes the transmission adaptation of audio and command streams, and improves cross-system data interaction efficiency. User interface adaptation module: This module automatically adjusts the interface layout and touch response logic based on system features, adapting to various screen forms such as notch screens and foldable screens, and resolving interaction lag issues. Virtualization Optimization Layer: Customized optimizations based on kernel-based virtual machines (KVM) and industry-standard container runtime management tools, including container startup optimization, resource isolation optimization, and hardware virtualization adaptation modules. Container Startup Optimization Module: Employs pre-loading core components and optimized image layering storage to shorten container startup time to under 10 seconds. Resource Isolation Optimization Module: Optimizes CPU and memory isolation strategies for cloud phone applications, avoiding resource contention during parallel application execution. Hardware Virtualization Adaptation Module: Deeply adapts to domestic GPUs, encoding accelerator cards, and other hardware to improve hardware resource virtualization utilization efficiency. Edge-Cloud Collaboration Layer: Focuses on data transmission and encoding / decoding optimization, including encoding / decoding optimization, transmission protocol optimization, and weak network defense modules. Encoding / Decoding Optimization Module: Utilizes H.265 encoding / decoding technology combined with adaptive bitrate adjustment to reduce audio and video streaming latency. Transmission Protocol Optimization Module: Customizes edge-cloud data transmission logic based on the Web Real-Time Communication (WebRTC) protocol to optimize the real-time transmission performance of command streams and audio streams. Weak Network Countermeasures Module: Reduces the number of buffering events and improves transmission stability in weak network environments through data compression and redundancy error correction technologies. Resource Scheduling Layer: Achieves precise matching between underlying resources and business needs, including a resource status monitoring module, an intelligent scheduling strategy module, and a multi-node collaboration module. Resource Status Monitoring Module: Collects real-time usage rates of underlying resources such as CPU, GPU, memory, and storage, generating resource status reports. Intelligent Scheduling Strategy Module: Dynamically adjusts resource allocation ratios based on resource status and business scenarios (e.g., gaming, office work), prioritizing resource supply for high-priority services.Multi-node collaboration module: Links four core nodes and 30 provincial nodes to achieve cross-node resource scheduling and improve performance support capabilities in high-concurrency scenarios.

[0025] It should be noted that this embodiment can perform targeted optimizations for different scenarios. Therefore, before optimization, it is necessary to detect the scenario to be optimized, specifically two scenarios: multi-system underlying adaptation optimization and high-concurrency scenario underlying performance optimization. During the operation of the cloud phone system, the scenario monitoring module deployed in the resource scheduling layer monitors the system's operating status and access requests in real time. When a new terminal is detected accessing the cloud phone system, scenario identification is triggered; when the number of concurrent connections exceeds a preset threshold (e.g., 1000 concurrent connections), or the application startup delay increases, or the number of audio / video streaming stutters increases, scenario identification is also triggered. The scenario monitoring module determines the type of scenario that needs optimization based on the trigger event type, system logs, and operating indicators: if it is a terminal access event, it is determined to be a "multi-system underlying adaptation optimization scenario"; if it is a concurrent pressure or performance degradation event, it is determined to be a "high-concurrency scenario underlying performance optimization scenario". This step is a prerequisite for the method of this embodiment, ensuring that subsequent optimization operations are accurately matched with the current requirements.

[0026] Step S20: When the current scenario to be optimized is multi-system underlying adaptation optimization, collect system characteristic information when the terminal accesses the system.

[0027] It should be noted that multi-system underlying adaptation and optimization refers to cloud phones connecting to different system terminals. System characteristic information includes operating system type, version, and underlying architecture characteristics. For example, system type (iOS 15, HarmonyOS 5.0, etc.), underlying architecture characteristics, screen type, etc. For specific procedures, please refer to... Figure 2 As shown, the optimization process is as follows: System feature acquisition. The system feature parsing module of the multi-system adaptation layer obtains information such as terminal system type (iOS 15, HarmonyOS 5.0, etc.), underlying architecture features, and screen type through the terminal access protocol.

[0028] Adaptation configuration generation. Based on the collected system characteristics, an adaptation configuration file is automatically generated, specifying the interaction parameters, interface layout rules, touch response thresholds, etc. of the end-to-end cloud collaboration SDK.

[0029] SDK Dynamic Adaptation. The edge-cloud collaboration SDK optimization module dynamically adjusts the SDK's transmission protocol and data parsing logic based on the configuration file to ensure compatibility with the underlying terminal system.

[0030] Interaction logic calibration. The interface interaction adaptation module adjusts the layout of interface elements and touch response latency based on configuration files to adapt to notch screens, foldable screens, and other form factors, avoiding interaction lag.

[0031] Adaptation effect verification. The system automatically detects the application's running status, interface response speed, and functional compatibility; if any abnormalities are found, it returns an error.

[0032] Once the configuration is readjusted and verification is successful, the adaptation is complete.

[0033] Specifically, when a user accesses the cloud phone via an iOS device (such as iPhone 15), a HarmonyOS device (such as Huawei Mate 60), or an Android device (such as Xiaomi 14), the system characteristic parsing module of the multi-system adaptation layer establishes a connection with the terminal through the terminal access protocol (such as HTTPS, WebSocket) and reads the device information reported by the terminal. The parsing module extracts the operating system type (such as iOS, HarmonyOS, Android), operating system version (such as iOS 17.2, HarmonyOS 5.0, Android 14), underlying architecture characteristics (such as iOS sandbox mechanism, HarmonyOS distributed soft bus characteristics, Android ART virtual machine characteristics), and screen form information (such as notch coordinates, foldable screen unfolding state, punch-hole screen position) from the user agent string, device fingerprint, or data reported by the dedicated SDK. For example, when an iPhone 15 is accessed, the parsing module obtains the iOS 17.2 system and its sandbox security mechanism, as well as the notch size and resolution of the Super Retina XDR screen; when a Huawei Mate 60 is accessed, it obtains HarmonyOS 5.0 and its distributed soft bus characteristics, and multi-window interaction characteristics. This information is encapsulated into a system feature data package and passed to the adaptation configuration generation module.

[0034] Step S30: Dynamically generate an adapted configuration file based on the system characteristic information.

[0035] In its implementation, the adaptation configuration generation module receives the system feature data package and then calls a pre-built multi-system adaptation rule library. This rule library contains adaptation templates for different system types and versions, such as sandbox bypass strategies for iOS, distributed service call parameters for HarmonyOS, and permission management configurations for Android. The module selects a basic template based on the collected system type and version, and then dynamically adjusts parameters according to the underlying architecture characteristics. For example, for iOS 17.2, the configuration file sets the SDK interaction parameters to "sandbox compatibility mode" and enables specific entitlements permission requests; the interface layout rules are set to adapt to the top safe area offset of the notch screen (e.g., increasing the status bar height by 44 pixels); and the touch response threshold is adjusted to adapt to the sensitivity parameters of iOS swipe gestures (e.g., expanding the edge return touch area width from 20 pixels to 30 pixels). For HarmonyOS 5.0, the configuration file enables distributed soft bus optimization parameters, the interface layout adapts to the dual-screen collaborative layout of the unfolded foldable screen, and the touch response supports latency compensation parameters for multi-window dragging. The configuration file uses JSON or XML format and contains a three-layer structure: SDK layer parameters (audio stream encoding format, command channel priority), interface layer parameters (control spacing, font scaling, safe area), and touch layer parameters (click latency compensation, scrolling smoothness coefficient, edge gesture recognition threshold). The configuration file includes the interaction parameters of the end-to-end cloud collaboration SDK, interface layout rules, and touch response thresholds.

[0036] Step S40: Adjust the transmission protocol and data parsing logic of the end-to-cloud collaboration SDK according to the configuration file, and calibrate the interface interaction logic to achieve multi-system underlying compatibility.

[0037] In its implementation, the edge-cloud collaboration SDK optimization module reads the configuration file generated in the above steps and dynamically loads the adaptation parameters. At the transmission protocol level, based on the system characteristics in the configuration file, the SDK adjusts the audio stream transmission protocol: for iOS systems, due to strict background restrictions, a WebRTC-optimized audio channel is used, with a longer keep-alive interval and audio session reconnection strategy configured; for HarmonyOS systems, the distributed soft bus API is called to achieve low-latency direct transmission of edge-cloud audio streams. At the data parsing logic level, the SDK dynamically switches parsers to address the differences in data formats across different systems: for example, instruction data received by iOS devices is parsed using Property List format, HarmonyOS devices use CBOR format, and Android devices use Protocol Buffers format. Meanwhile, the interface interaction adaptation module adjusts the cloud phone's screen rendering layer in real time based on the interface layout rules and touch response thresholds in the configuration file: for notch screens, it moves the cloud phone's status bar down by the corresponding pixels to avoid it being obscured by the notch; for foldable screens in unfolded state, it automatically switches to widescreen layout mode and increases the number of application icon columns; for touch response, it adjusts the touch event reporting frequency and prediction algorithm parameters according to the configuration, for example, enabling higher-precision touch prediction for iOS devices to compensate for the latency caused by its system touch scanning cycle. Through these adjustments, it ensures that the cloud phone can obtain a native-level interactive experience on different system terminals.

[0038] In one embodiment, the transmission protocol and data parsing logic of the cloud-edge collaboration SDK are adjusted according to the configuration file, and the interface interaction logic is calibrated. Specifically, this includes: dynamically adjusting the transmission adaptation parameters of the SDK's audio stream and command stream according to the configuration file, and adjusting the layout of interface elements and touch response latency to adapt to different screen forms.

[0039] In practical applications, for example, the edge-cloud collaboration SDK optimization module parses the "SDK layer parameters" section of the configuration file to obtain the transmission adaptation parameters for audio and command streams. For audio streams, parameters include encoding format (e.g., Opus, AAC), sampling rate (e.g., 48kHz, 24kHz), number of channels, bitrate range, and packet loss compensation strategy. For instance, when the configuration file detects that the terminal is an iOS device on a cellular network, the audio stream parameters are set to: AAC encoding format, 24kHz sampling rate, mono, 64kbps bitrate, and PLC (packet loss hiding) technology enabled; when the configuration file detects that the terminal is a HarmonyOS device connected to Wi-Fi, the parameters are set to: Opus encoding format, 48kHz sampling rate, stereo, 128kbps bitrate, and PLC disabled to reduce latency. The SDK reinitializes the audio encoder and transmission channels based on these parameters.

[0040] For the command stream, parameters include command compression algorithms (such as Snappy, Gzip), command merging strategies (such as a maximum merging delay of 50ms), and retransmission mechanisms. For example, the configuration file specifies that HarmonyOS devices use distributed soft bus direct command transmission, and the command merging delay is set to 20ms to improve response speed; it specifies that iOS devices use WebSocket long connections due to background restrictions, and the command merging delay is set to 100ms to reduce wake-up frequency and enable an exponential backoff retransmission strategy.

[0041] The interface interaction adaptation module simultaneously parses the "interface layer parameters" and "touch layer parameters" in the configuration file. For the layout of interface elements, the module dynamically adjusts the cloud phone's screen rendering and composition based on screen shape information (notch coordinates, foldable screen state, punch-hole screen position, and corner radius). Taking a foldable screen as an example, when the device is detected to be in an unfolded state, the module switches the cloud phone desktop from a single-column icon layout to a two-column layout, adjusts the application icon size to 120×120 pixels to fit the larger screen, and adjusts the positions of the status bar and navigation bar accordingly. For a notch screen, the module obtains the coordinates of the notch area (e.g., (0, 0) to (50, 200)) and moves the cloud phone's content rendering area down by 50 pixels to ensure that critical content is not obscured.

[0042] For touch response latency calibration, the module adjusts the touch event handling logic based on the touch thresholds in the configuration file. For example, if the configuration file sets "edge back touch area width = 30 pixels" for iOS devices, the module will prioritize a back gesture over an in-app click if the horizontal coordinate of the touch point is less than 30 pixels when it receives a touch event. If the configuration file sets "two-finger zoom sensitivity coefficient = 1.2" for foldable screen devices, the module will amplify the zoom factor when processing two-finger touches to compensate for the physical distance perception of large-screen operations. Simultaneously, the module adjusts the touch prediction algorithm parameters: enabling more aggressive touch prediction with a prediction step size of 2 frames for high refresh rate devices (such as 120Hz screens); and using conservative prediction for ordinary 60Hz devices to avoid accidental touches.

[0043] Furthermore, after adjusting the transmission protocol and data parsing logic of the end-to-cloud collaboration SDK according to the configuration file, and calibrating the interface interaction logic, this embodiment automatically detects the application running status, interface response speed, and functional compatibility. If any abnormality is found, the configuration file is regenerated until the adaptation verification is passed.

[0044] It's important to note that the adaptation verification module initiates an automated testing process upon completion. This module simulates common user operations (such as swiping the desktop, opening applications, returning to the homepage, and switching between multiple tasks), collecting three types of metrics: application running status (whether it crashes, number of ANRs), interface response speed (swiping frame rate, click response latency), and functional compatibility (whether camera access, file uploads, and clipboard synchronization are normal). For example, if the cloud desktop swiping frame rate on an iOS device is detected to be below 50fps, it is determined to be an interface response anomaly; if a distributed file transfer failure is detected on a HarmonyOS device, it is determined to be a functional compatibility anomaly. The verification module feeds back the anomaly information to the adaptation configuration generation module, which adjusts the configuration file parameters according to the anomaly type: if the swiping frame rate is low, the rendering resolution is reduced or the touch prediction algorithm is adjusted; if file transfer fails, the file transfer protocol is switched from HTTP to WebDAV or an alternative file chunking upload strategy is enabled. The adjusted configuration file is re-issued and executed until all verification items pass or the maximum number of retries (e.g., 3 times) is reached. This closed-loop verification mechanism ensures the reliability and stability of multi-system adaptation.

[0045] Step S50: When the current scenario to be optimized is the underlying performance optimization of a high-concurrency scenario, collect the status data of the underlying hardware resources of the cloud phone in real time.

[0046] In this embodiment, the specific process can be referred to Figure 3 As shown, the optimization process is as follows: Resource status monitoring. The resource status monitoring module collects underlying resource data such as CPU utilization, GPU load, memory usage, and network bandwidth of each node in real time, with a collection frequency of once every 10 seconds.

[0047] Business scenario identification. The system identifies the current business scenario (such as a cloud gaming scenario with 10,000 concurrent streams) and determines the business priority and resource requirements (such as high GPU load requirements).

[0048] Scheduling strategy decision-making. The intelligent scheduling strategy module combines resource status and business needs to decide on resource allocation schemes: prioritizing the allocation of GPU resources to high-priority businesses, adjusting CPU scheduling weights, and optimizing memory caching strategies.

[0049] Dynamic resource allocation. Through the resource isolation optimization module of the virtualization optimization layer, underlying resources are allocated according to the decision scheme to achieve resource isolation among multiple applications and avoid preemption.

[0050] Real-time performance adjustment. The encoding / decoding optimization module and transmission protocol optimization module of the edge-cloud collaboration layer synchronously adjust parameters to reduce streaming latency; if the resource status changes (such as CPU utilization exceeding 80%), the system returns to make a new decision to ensure stable performance.

[0051] In practical implementation, if the target is to optimize the underlying performance in high-concurrency scenarios, real-time status data of the underlying hardware resources of the cloud phone is collected. This status data includes the utilization rate of CPU, GPU, memory, and storage resources. For example, the resource status monitoring module of the resource scheduling layer collects resource usage data of all computing nodes at a preset frequency (e.g., 10 seconds / time, which can be dynamically adjusted to 5 seconds / time during high concurrency). For CPU, it collects the utilization rate of each physical core, average load, and run queue length; for GPU, it collects encoder utilization, video memory usage, and rendering load; for memory, it collects total memory, used memory, cache usage, and swap partition utilization; for storage, it collects IOPS, read / write latency, and disk queue depth; and simultaneously collects network bandwidth usage, TCP retransmission rate, and round-trip latency. The monitoring module aggregates this data to generate resource status reports, stores them in a time-series database, and marks the region to which each node belongs. For example, during peak cloud gaming hours (e.g., 8 pm), the Tianjin node may detect a GPU encoder utilization rate of 95%, a memory usage rate of 90%, and network egress bandwidth close to saturation; while the Guizhou node has a lighter load. This real-time data provides the basis for subsequent scheduling decisions.

[0052] Step S60: Identify the currently running business scenario and its priority.

[0053] In this embodiment, the business scenarios include cloud gaming, office applications, or high-concurrency streaming. The business identification module analyzes the application types and user behaviors of all cloud phone instances running on the current node. Business scenarios are identified through application process names, network traffic characteristics, and API call sequences: if a large number of instances are running game applications with high GPU load and high video stream bitrate, it is determined to be a "cloud gaming scenario"; if instances are running Office suites, WeChat Work, or email clients with frequent file read / write operations and stable CPU load, it is determined to be an "office application scenario"; if a large number of instances are conducting video conferencing, live streaming, or screen sharing with high network bandwidth usage and high audio / video encoding load, it is determined to be a "high-concurrency streaming scenario". The module assigns priorities to each scenario based on the business type and Service Level Agreement (SLA): for example, cloud gaming scenarios are latency-sensitive and are set to high priority; high-concurrency streaming scenarios have high bandwidth requirements and are set to medium priority; office application scenarios have relatively stable resource requirements and are set to normal priority. Simultaneously, the module also identifies different user levels within the same scenario (such as VIP users and regular users) to further refine the priority hierarchy.

[0054] Step S70: Dynamically allocate underlying hardware resources based on the resource status data and the business scenario priority.

[0055] In a specific implementation, underlying hardware resources are dynamically allocated based on the resource status data and the business scenario priority, including: determining a resource allocation scheme according to the business scenario priority, allocating GPU resources for high-priority businesses, adjusting CPU scheduling weights, and optimizing memory caching strategies; and isolating and allocating underlying resources according to the resource allocation scheme through the resource isolation module of the virtualization layer.

[0056] It's important to note that resource allocation schemes are determined based on the priority of business scenarios. For example, the scheduling module prioritizes resource needs for high-priority scenarios: for cloud gaming, dedicated GPU resources are allocated first, ensuring each game instance receives at least 1GB of video memory and 50% of the GPU encoder time slice; for high-concurrency streaming scenarios, shared GPU resources are allocated, but the encoding channel priority is guaranteed; for office applications, CPU and memory resources are primarily allocated, using CPU share (CFS quota) to ensure basic performance. For instance, when GPU resources are scarce at the Tianjin node, the scheduling module migrates cloud gaming instances to the Guizhou node or downgrades some non-critical game instances to CPU rendering. GPU resources are allocated to high-priority services, CPU scheduling weights are adjusted, and memory caching strategies are optimized. The scheduling module calls the resource isolation optimization module of the virtualization layer to achieve fine-grained allocation through cgroups and GPU virtualization technology: using CPU cgroups to set the CPU weight of high-priority services to 1024 (512 for normal services) to ensure that they get more time slices in CPU contention; using GPU SR-IOV or vGPU technology to allocate dedicated GPU virtualization functions to cloud gaming instances; in terms of memory management, locking memory pages for high-priority services to prevent them from being swapped to disk and increasing their page cache size.

[0057] Furthermore, the resource isolation module at the virtualization layer allocates underlying resources in isolation according to the resource allocation scheme. The resource isolation optimization module configures resource limits for different cloud phone instances based on the scheduling scheme. For example, cloud gaming instance A is configured with CPU bound to physical cores 0-3, memory limited to 8GB, and 2GB of reserved GPU memory; office application instance B is configured with a CPU share of 512, memory limited to 2GB, and uses a shared GPU but does not guarantee encoding latency. The isolation strategy also includes network bandwidth guarantees: bandwidth is reserved for high-priority services through TC (Traffic Control), and minimum guaranteed bandwidth and maximum burst bandwidth are set.

[0058] In this embodiment, to ensure the rationality of the resource allocation scheme, continuous monitoring of resource status changes is conducted. If resource utilization exceeds a preset threshold, the resource allocation decision is re-triggered, and the resource allocation scheme is dynamically adjusted. Specifically, the monitoring module continuously tracks resource usage. When it detects that the utilization of a certain type of resource exceeds a threshold (e.g., CPU utilization > 80% for 30 seconds; or GPU encoder queue length > 10), a dynamic adjustment process is triggered. The scheduling module reassesses the priority and actual resource consumption of all current instances and executes "preemptive scheduling": reallocating resources from low-priority services to high-priority services; or triggering "elastic scaling," scheduling some instances to nodes with lower loads. For example, when it is detected that the CPU utilization of the Tianjin node continuously exceeds 85% and cloud gaming latency increases, the scheduling module migrates some office application instances to the Henan node, releasing CPU resources for cloud gaming instances. At the same time, it adjusts the memory caching strategy, increasing the page cache of cloud gaming instances by 20% to reduce disk I / O.

[0059] Step S80: During the audio and video streaming process, dynamically adjust the encoding and decoding parameters and transmission protocol, and perform data compression and redundancy error correction operations in a weak network environment.

[0060] In this implementation, H.265 or AV1 encoding / decoding technology is used, combined with adaptive bitrate adjustment, to reduce audio and video streaming latency. Data transmission logic is customized based on the WebRTC protocol to optimize the real-time transmission performance of the command stream and audio stream. Specifically, the encoding / decoding optimization module and the transmission protocol optimization module of the end-to-cloud collaboration layer work together. The encoding / decoding optimization module uses H.265 or AV1 encoding / decoding technology, combined with an adaptive bitrate adjustment algorithm: based on the network bandwidth and allocated GPU resources collected in the above steps, the video encoding parameters are dynamically adjusted. When sufficient network bandwidth is detected, AV1 encoding is enabled and the resolution is increased to 1080P, with the bitrate set to 8Mbps; when network congestion or bandwidth decreases, H.265 encoding is switched to and the resolution is reduced to 720P, with the bitrate dynamically adjusted to 3Mbps, while the frame rate is adjusted from 60fps to 30fps. The transmission protocol optimization module customizes data transmission logic based on the WebRTC protocol: it optimizes the encapsulation strategy of audio and video RTP packets to reduce header overhead; it enables SVC (Scalable Video Coding) technology to prioritize the transmission of base layer data when network packet loss occurs; and it establishes independent QUIC connections for command streams and audio streams to provide lower transmission latency and better congestion control. For example, in weak network environments (packet loss rate > 5%), the module automatically enables NACK (Fast Retransmission) and FEC (Forward Error Correction) mechanisms to add redundant data to critical I-frames, reducing video stuttering.

[0061] In one embodiment, data compression and redundancy correction operations are performed in weak network environments. For example, the weak network adversarial module is activated when it detects a decline in network quality (e.g., RTT > 200ms, packet loss rate > 3%). The module employs a two-level optimization: The first level is data compression, using LZ4 or Zstandard algorithms to compress non-real-time data (e.g., file transfer, log reporting) to reduce the amount of data transmitted; for audio and video streams, more aggressive quantization parameters are enabled at the encoding level to reduce the bitrate. The second level is redundancy correction, using fountain codes or Reed-Solomon erasure coding technology to fragment audio and video data and add redundant data packets, so that the receiving end can recover the original data even if some data packets are lost. For example, when it is detected that a user is switching from a 5G network to a 4G network, causing a decrease in bandwidth, the module immediately reduces the video bitrate from 6Mbps to 2Mbps and enables 20% FEC redundancy to ensure smooth video playback; for file upload operations performed by the user, the module fragments and compresses the file before transmission and enables the TCP BBR congestion control algorithm to optimize throughput.

[0062] Furthermore, this embodiment can also coordinate multiple core nodes and branch nodes for cross-node resource scheduling when a single node's resources are insufficient. For example, the multi-node collaboration module maintains a global resource view, summarizing the real-time resource status of the four core nodes and 30 provincial nodes. When local node resources are insufficient and performance requirements cannot be met through internal adjustments, cross-node scheduling is triggered. The scheduling decision considers three factors: the remaining resource capacity of the target node, the proximity of the user's geographical location, and the cost of business data synchronization. For example, when the CPU and GPU of the Tianjin node both reach 95% utilization, the module will directly schedule newly connected cloud gaming instances to the Henan node; for already running instances, "hot migration" technology is used to smoothly switch after synchronizing memory status and disk data to the Henan node, without the user's awareness. Cross-node scheduling also supports "edge-core collaboration": for latency-sensitive services (such as cloud gaming), priority is given to scheduling to provincial edge nodes; for computationally intensive services (such as video transcoding), scheduling is done to core nodes. The module uses a global load balancing algorithm to achieve a balanced distribution of resource utilization among nodes, improving the overall system capacity.

[0063] In this embodiment, the current scenario to be optimized is detected. If it is a multi-system underlying adaptation optimization, terminal system characteristic information is collected, and a configuration file containing end-to-cloud collaboration SDK interaction parameters and interface layout rules is dynamically generated. Based on this, the SDK transmission protocol, data parsing logic, and interface interaction logic are adjusted to achieve multi-system underlying compatibility. If it is a high-concurrency scenario underlying performance optimization, underlying hardware resource status data is collected in real time, business scenarios and priorities are identified, underlying hardware resources are dynamically allocated based on resource status and priority, and encoding / decoding parameters and transmission protocols are dynamically adjusted in audio and video streams. Data compression and redundancy correction are performed under weak network conditions. The above method improves multi-system compatibility, high-concurrency resource utilization efficiency, and end-to-cloud collaborative transmission performance through scenario-adaptive optimization strategies.

[0064] Furthermore, this embodiment of the invention also proposes a storage medium storing a cloud phone underlying performance optimization program, which, when executed by a processor, implements the steps of the cloud phone underlying performance optimization method described above.

[0065] Reference Figure 4 , Figure 4 This is a structural block diagram of the first embodiment of the cloud phone underlying performance optimization system of the present invention.

[0066] like Figure 4 As shown, the cloud phone underlying performance optimization system proposed in this embodiment of the invention includes: Scene detection module 10 is used to detect the current scene to be optimized; The multi-system adaptation and optimization module 20 is used to collect system characteristic information when the terminal accesses the system when the current scenario to be optimized is multi-system underlying adaptation and optimization. The system characteristic information includes operating system type, version and underlying architecture characteristics. The multi-system adaptation and optimization module 20 is used to dynamically generate an adaptation configuration file based on the system characteristic information. The configuration file includes the interaction parameters, interface layout rules and touch response thresholds of the end-to-cloud collaboration SDK. The multi-system adaptation and optimization module 20 is used to adjust the transmission protocol and data parsing logic of the end-to-cloud collaboration SDK according to the configuration file, and to calibrate the interface interaction logic to achieve multi-system underlying compatibility. The high-concurrency performance optimization module 30 is used to collect the status data of the underlying hardware resources of the cloud phone in real time when the current scenario to be optimized is the underlying performance optimization of a high-concurrency scenario. The status data includes the utilization rate of CPU, GPU, memory and storage resources. The high-concurrency performance optimization module 30 is used to identify the currently running business scenario and the priority of the business scenario, which includes cloud gaming, office applications, or high-concurrency streaming. The high-concurrency performance optimization module 30 is used to dynamically allocate underlying hardware resources based on the resource status data and the business scenario priority. The high-concurrency performance optimization module 30 is used to dynamically adjust the encoding and decoding parameters and transmission protocol during audio and video streaming, and to perform data compression and redundancy error correction operations in weak network environments.

[0067] In this embodiment, the current scenario to be optimized is detected. If it is a multi-system underlying adaptation optimization, terminal system characteristic information is collected, and a configuration file containing end-to-cloud collaboration SDK interaction parameters and interface layout rules is dynamically generated. Based on this, the SDK transmission protocol, data parsing logic, and interface interaction logic are adjusted to achieve multi-system underlying compatibility. If it is a high-concurrency scenario underlying performance optimization, underlying hardware resource status data is collected in real time, business scenarios and priorities are identified, underlying hardware resources are dynamically allocated based on resource status and priority, and encoding / decoding parameters and transmission protocols are dynamically adjusted in audio and video streams. Data compression and redundancy correction are performed under weak network conditions. The above method improves multi-system compatibility, high-concurrency resource utilization efficiency, and end-to-cloud collaborative transmission performance through scenario-adaptive optimization strategies.

[0068] This application embodiment also provides a cloud phone underlying performance optimization device, including a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other through the communication bus. The memory is used to store cloud phone underlying performance optimization programs. When the processor executes the programs stored in the memory, it implements the above-mentioned cloud phone underlying performance optimization method.

[0069] The communication bus mentioned in the aforementioned cloud phone underlying performance optimization device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc.

[0070] The communication interface is used for communication between the aforementioned cloud phone underlying performance optimization device and other devices.

[0071] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0072] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0073] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0074] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0075] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0076] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0077] It should be understood that the above are merely illustrative examples and do not constitute any limitation on the technical solutions of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any restrictions on this.

[0078] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this invention. In practical applications, those skilled in the art can select some or all of the workflow to achieve the purpose of this embodiment according to actual needs, and no restrictions are imposed here.

[0079] In addition, for technical details not described in detail in this embodiment, please refer to the cloud phone underlying performance optimization method provided in any embodiment of the present invention, which will not be repeated here.

[0080] Furthermore, it should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0081] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0082] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0083] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

[0084] It is understood that the system provided in the embodiments of the present invention corresponds to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

Claims

1. A method for optimizing the underlying performance of cloud phones, characterized in that, The underlying performance optimization method for cloud phones includes: Detect the current scene that needs optimization; When the current scenario to be optimized is multi-system underlying adaptation optimization, system characteristic information is collected when the terminal accesses the system. The system characteristic information includes operating system type, version, and underlying architecture characteristics. Based on the system characteristic information, an adaptive configuration file is dynamically generated. The configuration file includes the interaction parameters, interface layout rules and touch response thresholds of the end-to-cloud collaboration SDK. Adjust the transmission protocol and data parsing logic of the end-to-cloud collaboration SDK according to the configuration file, and calibrate the interface interaction logic to achieve multi-system underlying compatibility; When the current scenario to be optimized is the underlying performance optimization of a high-concurrency scenario, the status data of the underlying hardware resources of the cloud phone is collected in real time. The status data includes the utilization rate of CPU, GPU, memory and storage resources. Identify the currently running business scenario and its priority, including cloud gaming, office applications, or high-concurrency streaming; Dynamically allocate underlying hardware resources based on the resource status data and the priority of the business scenario; During audio and video streaming, the encoding and decoding parameters and transmission protocol are dynamically adjusted, and data compression and redundancy error correction operations are performed in weak network environments.

2. The cloud phone underlying performance optimization method as described in claim 1, characterized in that, The step of adjusting the transmission protocol and data parsing logic of the cloud-edge collaboration SDK according to the configuration file, and calibrating the interface interaction logic, includes: The SDK's audio and command stream transmission adaptation parameters are dynamically adjusted according to the configuration file, as well as the layout of interface elements and touch response latency, to adapt to different screen forms.

3. The cloud phone underlying performance optimization method as described in claim 1, characterized in that, After the steps of adjusting the transmission protocol and data parsing logic of the end-to-cloud collaboration SDK according to the configuration file and calibrating the interface interaction logic, the method further includes: It automatically detects the application's running status, interface response speed, and functional compatibility. If any abnormalities are found, it returns to regenerate the configuration file until the adaptation verification is passed.

4. The cloud phone underlying performance optimization method as described in claim 1, characterized in that, The dynamic allocation of underlying hardware resources based on the resource status data and the business scenario priority includes: Determine the resource allocation scheme based on the priority of business scenarios, allocate GPU resources to high-priority businesses, adjust CPU scheduling weights, and optimize memory caching strategies; The resource isolation module of the virtualization layer isolates and allocates underlying resources according to the resource allocation scheme.

5. The cloud phone underlying performance optimization method as described in claim 4, characterized in that, The method further includes: Continuously monitor changes in resource status. If resource utilization exceeds a preset threshold, re-trigger resource allocation decisions and dynamically adjust the resource allocation scheme.

6. The cloud phone underlying performance optimization method as described in claim 1, characterized in that, The dynamic adjustment of encoding / decoding parameters and transmission protocol includes: Employing H.265 or AV1 encoding / decoding technologies, combined with adaptive bitrate adjustment, reduces audio and video streaming latency; We have customized the data transmission logic based on the WebRTC protocol to optimize the real-time transmission performance of command streams and audio streams.

7. The cloud phone underlying performance optimization method as described in claim 1, characterized in that, The method further includes: When a single node's resources are insufficient, multiple core nodes and sub-nodes are linked to perform cross-node resource scheduling.

8. A cloud phone underlying performance optimization system, characterized in that, The cloud phone underlying performance optimization system is applied to the cloud phone underlying performance optimization method as described in any one of claims 7 to 7, the system comprising: The scene detection module is used to detect the current scene to be optimized. The multi-system adaptation and optimization module is used to collect system characteristic information when the terminal accesses the system when the current optimization scenario is multi-system underlying adaptation and optimization. The system characteristic information includes operating system type, version and underlying architecture characteristics. The multi-system adaptation and optimization module is used to dynamically generate an adaptation configuration file based on the system characteristic information. The configuration file includes the interaction parameters, interface layout rules, and touch response thresholds of the end-to-cloud collaboration SDK. The multi-system adaptation and optimization module is used to adjust the transmission protocol and data parsing logic of the end-to-cloud collaboration SDK according to the configuration file, and to calibrate the interface interaction logic to achieve multi-system underlying compatibility. The high-concurrency performance optimization module is used to collect the status data of the underlying hardware resources of the cloud phone in real time when the current scenario to be optimized is the underlying performance optimization of a high-concurrency scenario. The status data includes the utilization rate of CPU, GPU, memory and storage resources. The high-concurrency performance optimization module is used to identify the currently running business scenario and the priority of the business scenario, which includes cloud gaming, office applications, or high-concurrency streaming. The high-concurrency performance optimization module is used to dynamically allocate underlying hardware resources based on the resource status data and the priority of the business scenario; The high-concurrency performance optimization module is used to dynamically adjust the encoding and decoding parameters and transmission protocol during audio and video streaming, and to perform data compression and redundancy error correction operations in weak network environments.

9. A cloud phone underlying performance optimization device, characterized in that, The cloud phone underlying performance optimization device includes: a memory, a processor, and a cloud phone underlying performance optimization program stored on the memory and executable on the processor. The cloud phone underlying performance optimization program is configured to implement the steps of the cloud phone underlying performance optimization method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a cloud phone underlying performance optimization program, which, when executed by a processor, implements the steps of the cloud phone underlying performance optimization method as described in any one of claims 1 to 7.

Images