Host Selection Algorithm Memory Sharing History Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Virtualization systems face challenges in optimizing VM placement to maximize memory sharing among virtual machines, leading to inefficient resource utilization and suboptimal host selection for live migration and initial placement.
Innovation Solution
A host selection algorithm that considers shared memory history to prioritize candidate hosts with higher memory sharing history, adjusting scores based on the average memory sharing history between the target VM and currently running VMs on candidate hosts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a standard host selection algorithm is used for VM placement, then the placement process is simple and fast, but memory sharing optimization is not achieved leading to inefficient resource utilization
Solution Approach 1:
The system pre-calculates and stores memory sharing history values for VMs before placement decisions are made. This historical data is maintained in a database, allowing the selection algorithm to quickly reference pre-computed sharing patterns without performing complex real-time analysis, thus achieving memory sharing optimization without excessive computational complexity during placement
Solution Approach 2:
The patent replaces complex real-time memory analysis with a simplified algorithm that uses pre-computed historical data. Instead of performing sophisticated memory sharing calculations at placement time, the system substitutes this with a straightforward score adjustment mechanism that adds a memory sharing component to the selection algorithm, making the system both efficient and easy to implement
2Loss of energy
If memory sharing optimization is implemented in the host selection algorithm, then resource efficiency improves, but the algorithm complexity increases
Solution Approach 1:
The system optimizes memory utilization by dynamically adjusting the selection criteria parameter - specifically by adding a memory sharing history component to the host selection score. This parameter change allows the algorithm to favor hosts with VMs that have high memory sharing potential, improving memory efficiency without requiring a complete redesign of the selection algorithm
Solution Approach 2:
The patent uses historical memory sharing data as a copy of past behavior patterns to predict future sharing potential. By replicating and analyzing historical sharing patterns rather than performing complex real-time memory analysis, the system achieves optimization with manageable algorithmic complexity
3Reliability
If the host selection algorithm analyzes memory sharing history for all candidate hosts, then optimal VM placement is achieved, but the computational overhead and time consumption increase
Solution Approach 1:
The system performs memory sharing history analysis in advance and stores the results in a database. When a VM placement decision is needed, the pre-computed historical data is immediately retrieved and used to adjust selection scores, eliminating the need for time-consuming real-time analysis and enabling fast placement decisions with high optimality
Solution Approach 2:
The system dynamically balances between thorough analysis and quick decision-making by using pre-computed historical data that can be quickly retrieved and adjusted. The algorithm adapts by adding a memory sharing component to the selection process without requiring complete re-analysis of all candidate hosts, thus maintaining placement optimality while reducing time consumption
Data Source
AI summary
A mechanism for shared memory history optimization in a host selection algorithm for VM placement is disclosed. A method of embodiments of the invention includes determining candidate hosts to place a target virtual machine (VM), obtaining memory sharing history of the target VM with one or more VMs hosted by each of the candidate hosts, determining an average memory sharing history amount for each of an optimized number of the candidate hosts based on the obtained memory sharing history of the target VM with one or more VMs hosted by each of the optimized number of candidate hosts, and adjusting a score in a general selection algorithm for a candidate host with the highest average memory sharing history amount.