Network Cache Deduplication Analytics for VM Load Balancing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current load-balancing in virtualized network environments fails to maximize caching performance due to the lack of analytics in resource utilization assessments, leading to sub-optimal virtual machine allocation and increased network traffic.

Innovation Solution

A system and method for analyzing network traffic patterns to consolidate virtual machines with similar traffic patterns onto the same host, utilizing packet inspection and caching analytics to re-allocate VMs and increase cache usage efficiency, thereby maximizing caching and reducing network traffic.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional load-balancing operations are performed using only absolute resource utilization values, then the load distribution is simple to implement, but caching performance is not maximized and network traffic is not optimized

Engineering Contradiction:
Improvecaching performanceVSAvoidload-balancing complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system performs preliminary analytics on network traffic patterns before making load-balancing decisions. By analyzing traffic patterns in advance and identifying VMs with similar patterns, the system prepares optimization opportunities before workload migration, enabling caching performance improvement without reactive complexity

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms by continuously monitoring network traffic patterns and using this information to inform load-balancing decisions. The analytics component provides feedback on traffic similarity metrics that guide the placement of VMs on hosts to maximize caching efficiency

Inventive Principle:
Principle #23Feedback

2Reliability

If virtual machines with similar network traffic patterns are consolidated onto common hosts, then cache usage efficiency increases, but the virtual machine allocation becomes more complex

Engineering Contradiction:
Improvecache usage efficiencyVSAvoidVM allocation complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary analysis of network traffic patterns to identify VMs with similar characteristics before consolidation. This advance preparation enables the system to make informed decisions about which VMs to place together on hosts, maximizing cache efficiency without random or reactive allocation

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes the allocation parameters by incorporating network traffic pattern similarity metrics into the VM placement decision process. By using traffic pattern analysis as an additional parameter beyond traditional resource utilization, the system achieves more efficient cache usage while maintaining manageable complexity

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If analytics operations are performed on network traffic patterns, then VM consolidation accuracy improves, but the computational overhead increases

Engineering Contradiction:
Improvetraffic pattern matching accuracyVSAvoidcomputational overhead
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system extracts only the essential characteristics of network traffic patterns that are relevant for caching optimization, rather than analyzing all aspects of traffic. By focusing on key pattern metrics, the system achieves sufficient accuracy for VM consolidation while minimizing unnecessary computational overhead

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system applies analytics selectively to identify local similarities in traffic patterns among VMs that are candidates for consolidation. Rather than performing exhaustive analysis on all VMs in the data center, the system focuses computational resources on relevant subsets, improving accuracy where needed while reducing overall overhead

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS10447800B2Network cache deduplication analytics based compute cluster load balancer
Publication Date: 2019.10.15 DELL PROD LP
  • US10447800B2 patent drawing
  • US10447800B2 patent drawing
  • US10447800B2 patent drawing

AI summary

A system, method, and computer-readable medium for performing a virtualized network environment cache control operation, comprising: analyzing network traffic of a plurality of virtual machines executing on a plurality of hosts within the virtualized network environment, each of the plurality of hosts comprising a network cache; identifying whether more than one virtual machine includes a similar network traffic pattern; and, configuring a first virtual machine and a second virtual machine to executing on a common host, the first virtual machine and the second virtual machine having the similar network pattern.