Virtual Machine Memory Distribution via Software Similarity Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large server cluster systems, the effectiveness of memory distribution using page sharing technology is hindered by the low similarity of software running on virtual machines, leading to inefficient memory usage and resource wastage, as it relies heavily on the similarity of software across virtual machines.
Innovation Solution
A system and method that collects software type information from virtual machines across servers, forms a tree structure based on similarities, and migrates virtual machines with similar software types to optimize memory distribution, ensuring better page sharing and resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If page sharing technology is used for memory distribution in large server cluster systems, then memory distribution can be achieved, but the effectiveness is hindered by low software similarity across virtual machines, leading to poor memory sharing effect and resource wastage
Solution Approach 1:
The patent implements dynamic migration of virtual machines between servers based on real-time software type analysis. The system continuously monitors software characteristics and automatically migrates VMs to optimize software similarity on each server, making the memory sharing effect dynamic rather than static. This resolves the contradiction by adapting the system configuration to maximize page sharing effectiveness without requiring pre-existing software similarity.
Solution Approach 2:
The patent introduces a cluster management module as an intermediary that analyzes software types and coordinates VM migration across servers. This intermediary component enables the system to overcome the limitation of low software similarity by actively managing and redistributing VMs to create optimal grouping, thereby improving memory distribution effectiveness without requiring inherent software similarity.
2Productivity
If virtual machines with dissimilar software are distributed across servers using page sharing technology, then more virtual machines can be hosted, but memory resources are wasted and fewer virtual machines can effectively run in the system
Solution Approach 1:
The system dynamically adjusts VM placement to optimize memory sharing while maintaining high productivity. By continuously monitoring software similarity and migrating VMs as needed, the system ensures that memory resources are effectively shared among similar VMs, reducing waste while still hosting a large number of VMs across the cluster.
Solution Approach 2:
The patent implements a feedback mechanism where the cluster management module continuously analyzes software types and migration effects. This feedback loop allows the system to learn from past migrations and optimize future placements, ensuring that memory resources are not wasted while maintaining high VM density and productivity in the cluster system.
3Quantity of substance
If virtual machines are migrated to optimize software similarity, then memory distribution efficiency is enhanced, but system complexity increases due to migration management
Solution Approach 1:
The cluster management module performs multiple functions: it analyzes software types, decides migration timing, executes migrations, and monitors results. By consolidating these diverse functions into a single multi-functional module, the system enhances memory distribution efficiency while managing complexity through functional integration rather than proliferation of separate components.
Solution Approach 2:
The system implements self-service through automated software type analysis and migration decision-making. The cluster management module autonomously identifies migration opportunities based on software similarity and executes migrations without requiring manual intervention, thereby improving memory distribution efficiency while keeping operational complexity manageable through automation.
Data Source
AI summary
Disclosed are a system and a method for managing excessive distribution of memory. Based on a page sharing technology, the software types of virtual machines running in respective servers in a cluster are collected, and the virtual machines with similar running software types are migrated from a server to a specified server, so that the page sharing effect of the virtual machines and the excessive distribution effect of memory are better, the bearing capability of the servers in the system is ensured not to be wasted, the utilization rates of memory and resources of the whole system are combined optimally, and the memory of the whole cluster system is distributed better; moreover, a fewer number of servers run, the energy and the running cost are saved, less pressure is caused to the environment, the emission of carbon dioxide is reduced, therefore, the disclosure has a great social effect and economic effect.