Virtual Machine Deployment via Data Fragmentation and Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtualization methods lead to an excessively high load on virtual machines due to large amounts of data being deployed on a single machine, causing inefficiencies and potential overloading.
Innovation Solution
The method involves scrambling and fragmenting data into smaller nodes, allocating a target virtual machine based on the data amount of each node, and deploying these nodes to prevent overloading by matching the virtual machine's processing capability, thereby achieving load balancing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple flagship stores are deployed onto one virtual machine to increase resource utilization, then resource utilization rate is improved, but the virtual machine becomes overloaded and system reliability deteriorates
Solution Approach 1:
The patent segments data by store type (flagship stores vs. ordinary stores) and deploys them to different virtual machines. Flagship store data is deployed to dedicated virtual machines while ordinary store data is deployed to shared virtual machines, preventing overload on any single machine and maintaining system reliability while still achieving resource utilization optimization.
2Productivity
If cores in a low-load state are shared to create more virtual machines, then resource utilization rate is improved, but the virtual machine load increases excessively
Solution Approach 1:
The patent applies different resource allocation strategies to different types of stores. Flagship stores receive dedicated virtual machine resources ensuring stable performance, while ordinary stores share virtual machine resources to maximize utilization. This localized quality approach optimizes resource usage without causing excessive load on any single virtual machine.
Data Source
AI summary
The present disclosure provides virtual machine deployment methods and apparatuses. One exemplary virtual machine deployment method comprises: acquiring a fragment node by scrambling and fragmenting to-be-processed data; allocating a target virtual machine to the fragment node according to the data amount of the fragment node; and deploying the fragment node onto the target virtual machine. According to some embodiments of the present disclosure, when a virtual machine is allocated to a fragment node, the data amount of the fragment node can be controlled firstly by scrambling data. Then a virtual machine matching the data amount can be allocated according to the actual data amount of the fragment node, so as to prevent the virtual machine from overloading, thereby achieving better load balancing.