Virtual Machine Traffic Rate Estimation for Host Deployment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies lack an effective method for accurately estimating traffic rates between virtual machines, leading to inefficient network bandwidth usage and increased service delays due to suboptimal deployment of virtual machine pairs across physical hosts.
Innovation Solution
A method and apparatus for estimating traffic rates between virtual machines by obtaining and analyzing traffic parameter sets, including rates of sending and receiving traffic across multiple physical hosts and switching devices, to accurately calculate traffic rates using specific formulas that account for traffic flows and correlations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If virtual machine pairs are deployed on different physical hosts, then network bandwidth resources are occupied and service delay increases, but deployment flexibility and resource utilization are improved
Solution Approach 1:
The patent estimates traffic rates between virtual machine pairs in advance before deployment decisions are made. By calculating the traffic rate Tvm-x1→vm-y1 using monitoring data from switching devices and physical hosts, the system can predict which virtual machine pairs will generate high traffic and should be deployed on the same physical host, thus avoiding future service delays while maintaining deployment flexibility
2Loss of time
If virtual machine pairs with high traffic rates are deployed on the same physical host, then service delay is reduced, but accurate traffic rate estimation becomes more complex
Solution Approach 1:
The patent segments the traffic rate estimation into manageable components by monitoring specific parameters at different levels: virtual machine level (sending and receiving traffic rates), physical host level (outgoing and incoming traffic rates), and switching device level (traffic rates between switching devices). This segmentation allows the complex estimation problem to be solved by combining simpler measurements from multiple sources
Solution Approach 2:
The patent introduces switching devices as intermediaries to facilitate traffic rate estimation. The switching devices monitor and report traffic rates between virtual machines, physical hosts, and other switching devices, providing the necessary data for accurate estimation without requiring direct access to all virtual machine communications
3Measurement precision
If traffic monitoring is implemented across multiple switching devices and physical hosts, then traffic rate estimation accuracy is improved, but system complexity and data collection overhead increase
Solution Approach 1:
The patent makes the switching devices multi-functional by having them perform both their traditional switching function and additional traffic monitoring function. The switching devices collect and report traffic rates not only for their direct connections but also for connections through other switching devices, enabling comprehensive traffic rate estimation without adding separate dedicated monitoring hardware
Data Source
AI summary
An apparatus and a method for estimating a traffic rate between a virtual machine pair includes, when a rate of traffic sent by a virtual machine vm-x1 to a virtual machine vm-y1 is estimated, obtaining at least rates of sending traffic by N21 virtual machines deployed in N2 physical hosts and including the virtual machine vm-x1, rates of traffic sent by N1 switching devices to N4 switching devices, rates of receiving traffic of N31 virtual machines deployed in N3 physical hosts and including the virtual machine vm-y1, and rates of outgoing traffic of the N4 switching devices.