Layered Virtual Machine QoS Control With Token-Bucket Queues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud computing platforms face challenges in efficiently managing and controlling the quality of service (QoS) for virtual machines, particularly in large-scale environments where multiple user virtual machines send operation requests simultaneously, leading to resource inefficiencies and potential network congestion.
Innovation Solution
A method and apparatus utilizing a token bucket mechanism to control QoS by generating tokens periodically based on an upper limit, storing operation requests in a target queue when tokens are insufficient, and deducting tokens when sufficient, ensuring requests are processed only when the required token quantity is met, with separate control layers for entire machine, virtual machine, and block device Bdev layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If multiple user virtual machines send operation requests simultaneously in a large-scale cloud computing environment, then the computing power and storage capabilities are enhanced, but resource inefficiencies and network congestion occur
Solution Approach 1:
The patent segments the virtual machine cluster into multiple layers (first layer, second layer, third layer) with separate token buckets for each layer. This segmentation allows independent QoS control at different hierarchical levels, enabling fine-grained resource management that reduces waste while maintaining high computing power across the entire system.
Solution Approach 2:
The patent implements preliminary action by pre-generating tokens in token buckets before operation requests are processed. Tokens are generated in advance based on QoS policies, allowing the system to make quick decisions about request admission without real-time calculation delays, thus improving resource utilization efficiency.
2Power
If multiple user virtual machines send operation requests simultaneously, then the computing power is enhanced, but network congestion occurs
Solution Approach 1:
The token bucket mechanism performs preliminary action by pre-calculating and storing tokens before request processing. When operation requests arrive, the system immediately checks available tokens and makes admission decisions instantly, preventing request queues from building up and avoiding network congestion before it occurs.
Solution Approach 2:
The patent implements feedback mechanisms where the QoS management module continuously monitors token consumption and adjusts token generation rates accordingly. This feedback loop ensures that the system adapts to actual load conditions, dynamically controlling request admission to prevent network congestion while maintaining high computing power.
3Loss of energy
If a token bucket mechanism is used to control QoS, then resource utilization is enhanced, but the system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the QoS control system into multiple independent token buckets, each associated with a specific layer. This modular approach makes the complex QoS management system more manageable, as each token bucket can be controlled and monitored independently, reducing the overall system complexity while improving resource utilization.
Solution Approach 2:
The token bucket mechanism implements self-service by automatically managing its own state through periodic token generation and consumption tracking. The system self-regulates QoS without requiring external intervention, simplifying operation while improving resource utilization efficiency.
4Speed
If operation requests are processed immediately without token checking, then the transmission speed is improved, but QoS control is lost
Solution Approach 1:
The patent implements preliminary action by pre-checking token availability before processing operation requests. This advance verification ensures that QoS control is maintained while allowing rapid processing of legitimate requests, as the token checking happens in advance rather than during the critical processing path.
Solution Approach 2:
The token bucket mechanism provides self-service QoS control by automatically verifying token availability and making admission decisions without external intervention. This self-regulating mechanism maintains reliable QoS control while enabling fast request processing, as the system autonomously manages the balance between speed and control.
Data Source
AI summary
A method for controlling quality of service of a virtual machine, an electronic device and a storage medium are provided, relating to the fields of cloud computing, virtualization, big data and other technologies. The method includes: storing an operation request in a queue corresponding to a layer of a virtual machine cluster when an available token quantity in a token bucket of the layer is unable to meet a target token quantity required for the operation request, tokens of the layer being periodically generated according to an upper limit of quality of service of the layer; deducting the target token quantity from an updated available token quantity when the available token quantity in the token bucket is updated to meet the target token quantity required for the operation request; and sending the operation request from the queue to the layer for processing when deducting the target token quantity successfully.