Cloud CPU Underclocking Risk Identification via Frequency Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Public cloud providers face challenges in ensuring isolation between tenants and maintaining Service Level Agreements due to CPU underclocking, which occurs when high-load services cause temperature increases leading to thermal design power thresholds, affecting performance and latency for other tenants sharing the same physical socket.
Innovation Solution
A method for identifying underclocking risks by collecting frequency fluctuations of CPU units and CPU utilization rates in a public cloud environment, allowing for the sifting out of risky virtual machines and subsequent migration to prevent performance degradation and ensure SLA compliance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple tenants share different physical cores on the same host to achieve flexible sale and on-demand use of resources, then resource utilization and flexibility are improved, but CPU underclocking occurs when high-load services cause temperature increases, leading to performance degradation and latency for other tenants
Solution Approach 1:
The patent segments the physical host into multiple CPU units, each containing multiple cores. By collecting frequency fluctuations at the CPU unit level rather than individual core level, the system can identify underclocking events affecting specific CPU units and migrate affected virtual machines to different CPU units, thereby isolating the impact and maintaining overall system reliability while preserving resource sharing flexibility.
2Reliability
If basic multi-tenant isolation and security assurance are provided with vendor virtual machine specifications of expected operating frequencies, then security and baseline performance are ensured, but existing solutions cannot rapidly and accurately identify risky virtual machines affected by underclocking
Solution Approach 1:
The patent implements a feedback mechanism by continuously collecting frequency fluctuations of CPU units and CPU utilization rates of tenant virtual machines. This real-time monitoring creates a feedback loop that enables the system to dynamically identify virtual machines experiencing underclocking events, transforming the previously static security assurance into an active detection and response system that can rapidly identify risky virtual machines.
Solution Approach 2:
The patent replaces manual or traditional monitoring methods with an automated electronic detection system that collects frequency fluctuation data and utilization rate data through software agents. This substitution of mechanical/manual processes with electronic automated systems enables rapid and accurate identification of risky virtual machines, overcoming the limitations of existing detection approaches.
3Reliability
If virtual machines are migrated to prevent performance degradation and ensure SLA compliance, then service level agreement compliance is improved, but the complexity of monitoring and managing frequency fluctuations and utilization rates increases
Solution Approach 1:
The patent creates a multi-functional monitoring system that simultaneously collects frequency fluctuation data, calculates CPU utilization rates, identifies underclocking events, and triggers virtual machine migration decisions. By consolidating these multiple functions into a single integrated system, the patent reduces overall complexity compared to having separate systems for each function, while ensuring SLA compliance through automated multi-functional operation.
Data Source
AI summary
A method for identifying underclocking risks in a public cloud, an electronic device, and a storage medium are provided. The method includes collecting frequency fluctuations of CPU units in a host in a public cloud environment; wherein each of the CPU units includes a plurality of cores; collecting CPU utilization rates of tenant virtual machines in the host; and sifting out risky virtual machines from the tenant virtual machines according to the frequency fluctuations of the CPU units and the CPU utilization rates of the tenant virtual machines.