Hypervisor Resource Allocation for Real-Time OS Interference Prevention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The phenomenon of system resources allocated according to original specifications becoming insufficient on a hypervisor due to some operating system domains excessively using resources, leading to inadequate operation of remaining domains, particularly real-time operating systems, which cannot meet real-time requirements.
Innovation Solution
An operating system performance interference preventing apparatus that includes a hardware physical layer, operating system layer, and hypervisor, which allocates resources by precisely measuring back-end and native device driver usage, ensuring accurate resource allocation for each domain, including application, front-end device driver, and basic resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the hypervisor allocates hardware resources for each operating system according to given specifications, then the hardware resources are efficiently used and multiple operating systems can operate simultaneously, but some operating system domains may excessively use hardware resources causing resource insufficiency for other operating system domains
Solution Approach 1:
The patent measures and calculates the hardware resource usage of device driver domains (including back-end device drivers and native device drivers) in advance before allocating resources to operating system domains. This preliminary measurement ensures that the hypervisor can allocate sufficient resources to real-time operating systems by knowing the actual resource consumption patterns of device drivers beforehand, preventing resource insufficiency during runtime.
Solution Approach 2:
The patent implements a feedback mechanism where the hypervisor continuously monitors hardware resource usage by operating system domains and adjusts resource allocation accordingly. When resource usage exceeds allocated specifications, the system can reallocate resources to ensure real-time operating systems maintain their required performance levels, creating a dynamic balance between productivity and reliability.
2Productivity
If a specific operating system excessively uses about 70% of hardware resources, then that operating system can perform intensive operations, but other operating systems cannot sufficiently use the hardware resources allocated according to specifications
Solution Approach 1:
The patent applies preliminary anti-action by measuring and accounting for device driver resource consumption before it can excessively impact operating system domains. The hypervisor calculates the hardware resource usage of device driver domains and pre-adjusts the allocation to operating system domains, preventing any single operating system from excessively consuming resources by establishing proper resource boundaries in advance.
Solution Approach 2:
The patent changes the resource allocation parameters by incorporating device driver domain resource usage measurements into the hypervisor's allocation calculations. Instead of allocating resources based solely on operating system specifications, the system adjusts allocation parameters to account for actual device driver consumption, ensuring that intensive operations by one system do not deprive others of their allocated resources.
3Ease of manufacture
If the hypervisor does not measure device driver resource usage, then the resource allocation process is simple and fast, but the allocated resources do not reflect actual resource consumption leading to resource insufficiency
Solution Approach 1:
The patent performs preliminary measurement of device driver resource usage during system initialization or configuration phases, separating the measurement process from the runtime allocation process. This allows the hypervisor to gather accurate resource consumption data from device driver domains in advance, then use this pre-measured information for allocation decisions, maintaining both measurement precision and allocation simplicity.
Solution Approach 2:
The patent introduces an intermediary measurement and calculation layer between the device driver domains and operating system domains. The hypervisor acts as this intermediary, measuring device driver resource usage, calculating the impact on available resources, and then allocating resources to operating system domains based on these calculations. This intermediary process ensures measurement accuracy while keeping the allocation mechanism manageable.
Data Source
AI summary
The present invention relates to an operating system performance interference preventing apparatus of a hypervisor system.The present invention includes a hardware physical layer, an operating system layer composed of a plurality of operating systems of different types, and a hypervisor that allocates basic resources for using system resources of the hardware physical layer for each of a plurality of operating system domains corresponding to a plurality of operating systems constituting the operating system layer so that the plurality of operating systems operate in a virtual machine environment, and the basic resources allocated by the hypervisor for each of the plurality of operating system domains include a first resource consumed when a specific operating system domain, which is one of the plurality of operating system domains, directly uses the system resources, a second resource consumed when the hypervisor uses the system resources, and a third resource consumed in a process of using the system resources by a device driver domain that drives an input/output device of the hardware physical layer when the specific operating system domain uses the input/output device.According to the present invention, there is an effect of solving the problem that a phenomenon, in which system resources allocated according to the original specifications for each operating system on the hypervisor become insufficient, occurs due to an occurrence of a situation, in which some operating system domains operating on the hypervisor excessively use system resources, and remaining operating system domains, particularly a real-time operating system, cannot operate while satisfying real-time requirements.