Virtual Machine Hardware Device Switching via Resource Converting Module
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual machine systems fail to accurately share real hardware devices among multiple operating systems, leading to inconsistencies in hardware representation, underutilization of hardware features, and performance bottlenecks due to intermediate processes.
Innovation Solution
A virtual machine system with a Service OS, Guest OSs, and a Virtual Machine Monitor (VMM) that includes a resource converting module, switching judging module, and access control module to manage resource allocation and device switching, allowing direct access to real hardware devices based on sharing modes such as all-time sharing, time division switching, and all-time exclusive modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a general virtual device platform is used to virtualize hardware devices, then multiple Guest OSs can access hardware through a Service OS and VMM, but the hardware devices displayed by the virtual machine system are not consistent with the real hardware devices and users cannot see the real hardware devices
Solution Approach 1:
The patent creates a virtual hardware platform that copies the structure and properties of real hardware devices. The virtual PCI bus and virtual hardware devices are designed to mirror the real hardware architecture, allowing Guest OSs to see and interact with hardware representations that accurately reflect the physical devices while maintaining virtualization benefits.
2Adaptability or versatility
If a general virtual device platform is used to virtualize hardware devices, then hardware can be accessed through VMM and Service OS, but some properties of the real hardware cannot be used, such as hardware accelerating property of graphical cards, new standard of USB, and advanced properties of sound cards
Solution Approach 1:
The patent implements dynamic device assignment that allows the virtualization system to adaptively allocate and configure hardware devices based on Guest OS requirements. The system can dynamically enable direct access to specific hardware devices and their advanced properties, allowing features like GPU hardware acceleration and USB 3.0 capabilities to be utilized when needed, while maintaining the virtualization architecture.
3Ease of operation
If a general virtual device platform is used to virtualize hardware devices, then hardware access is managed through intermediate processes, but this impacts the performance of accessing the hardware
Solution Approach 1:
The patent segments hardware access into different paths: standard virtualized access through the Service OS for general purposes, and direct access paths for performance-critical operations. The resource converting module is segmented to handle only the necessary address translations, minimizing intermediate processing steps while maintaining security and management controls.
Solution Approach 2:
The patent introduces a resource converting module as an intelligent intermediary that performs address space conversions between Guest OS virtual addresses and host physical addresses. This mediator enables direct hardware access by Guest OSs while maintaining the abstraction benefits of virtualization, eliminating the need for full I/O request interception and processing through the Service OS.
4Adaptability or versatility
If a general virtual device platform is used to virtualize hardware devices, then a virtual hardware platform is generated for the Guest OS, but the properties of the real hardware are not completely exploited due to no corresponding management and control
Solution Approach 1:
The patent implements feedback mechanisms where the VMM and Service OS monitor hardware device usage and performance metrics. Based on this feedback, the system dynamically adjusts resource allocation, device assignment, and virtualization parameters to optimize hardware exploitation. The system can detect when advanced hardware properties are available and configure access accordingly.
Data Source
AI summary
The present invention provides a virtual machine system and a method for switching hardware devices thereof. In the virtual machine system, the hardware devices are classified into all-time sharing devices, time division switching devices and all-time exclusive devices based on using fashions. For the time division switching devices, the method for switching hardware devices is adopted, comprising steps of: upon system switching, sending a corresponding switching signal to a device model (DM) module corresponding to a foreground Guest OS and a DM module corresponding to a background Guest OS; the DM module corresponding to the foreground Guest OS releasing the hardware devices from its device space; the DM module corresponding to the background Guest OS adding the hardware devices released by the DM module corresponding to the foreground Guest OS to its device space. It is possible to implement switching of hardware devices among different Guest OSs by the system and method of the invention.