Nonstop Computing Fabric via DNS Zone Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current virtualization systems are sub-optimal as they can lead to resource contention among partitions on a host system, resulting in bandwidth limitations and inadequate redundancy, security, and quality of service, especially in distributed computing environments.

Innovation Solution

A para-virtualization system architecture that includes infrastructure partitions for resource management and interconnect service partitions to manage zones, ensuring dedicated resources, redundancy, and secure communication across host platforms, using DNS services to define and manage zones and interconnectivity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If partitions share resources on the same host system, then resource utilization efficiency is improved, but resource contention and bandwidth limitations occur

Engineering Contradiction:
Improveresource utilization efficiencyVSAvoidbandwidth availability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent segments the computing fabric into multiple zones using DNS-based zone files, where each zone represents a isolated network segment. Partitions are assigned to specific zones, and interconnect service partitions manage zone boundaries. This segmentation allows resource sharing within zones while preventing cross-zone resource contention, resolving the contradiction between resource utilization efficiency and bandwidth availability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The interconnect service partition acts as an intermediary between different zones, managing communication and resource allocation across zone boundaries. It uses DNS services to resolve addresses and enforce access policies, enabling efficient resource sharing within zones while maintaining bandwidth guarantees through controlled inter-zone communication.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If partitions are designated as backups for one another, then data reliability is improved, but system complexity and management overhead increase

Engineering Contradiction:
Improvedata reliabilityVSAvoidsystem management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system implements self-service failover mechanisms where standby partitions automatically assume the roles of failed primary partitions without manual intervention. DNS zone files automatically update to reflect the new primary-standby relationships, and the interconnect service partition manages the failover process transparently. This reduces management overhead while maintaining reliability through automated backup and failover capabilities.

Inventive Principle:
Principle #25Self-service

3Adaptability or versatility

If partitions can intercommunicate freely, then system versatility is improved, but security between unauthorized partitions deteriorates

Engineering Contradiction:
Improvesystem versatilityVSAvoidsecurity risks
Core Design Contradiction:
Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The computing fabric is segmented into multiple isolated zones, each with its own DNS zone file defining permitted communication patterns. Partitions can only communicate with other partitions in the same zone or with explicitly authorized partitions in other zones. This segmentation maintains system versatility within zones while enforcing security boundaries between zones, preventing unauthorized intercommunication.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The interconnect service partition serves as a security intermediary that mediates all cross-zone communication. It enforces access control policies defined in DNS zone files, allowing versatile communication within authorized boundaries while blocking unauthorized access attempts. This intermediary approach preserves system versatility for authorized operations while mitigating security risks through centralized policy enforcement.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS10965616B2Nonstop computing fabric arrangements
Publication Date: 2021.03.30 UNISYS CORP
  • US10965616B2 patent drawing
  • US10965616B2 patent drawing
  • US10965616B2 patent drawing

AI summary

Systems and methods for non-stop computing in a virtualization fabric are disclosed. One system includes a computing fabric comprising a plurality of host platforms, the plurality of host platforms including at least a first host platform and a second host platform communicatively connected to the first host platform. The system also includes an interconnect service partitions residing on the first host platform. The system includes a plurality of guest partitions distributed across the plurality of host platforms. The system further includes a DNS server instance managed by at least one of the plurality of interconnect service partitions and defining at least one zone, the at least one zone including one or more partitions from among the plurality of guest partitions distributed across the plurality of host platforms.