Distributed OS Virtual Resource Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtualization technologies fail to efficiently provision resources across multiple physical machines, leading to challenges in virtual machine migration, service delays, and reduced capability due to tight binding with underlying hardware resources, especially across geographically distributed data centers.
Innovation Solution
A multi-layer architecture with a distributed operating system layer that aggregates computing resources from multiple physical machines to provide virtualized resources, allowing dynamic provisioning and re-provisioning of virtual machines across geographically distributed data centers, decoupling virtual machines from specific physical machine resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If virtual machines are tightly bound to underlying hardware resources of multiple physical machines, then resource provisioning is simplified, but migration capability and fault tolerance deteriorate
Solution Approach 1:
The patent segments the binding between virtual machines and physical hardware by introducing a distributed operating system layer that abstracts physical resources. Virtual machines are divided into virtualized resource units that can be independently allocated and migrated across different physical machines without maintaining tight hardware binding.
Solution Approach 2:
The distributed operating system acts as an intermediary layer between virtual machines and physical hardware resources. This mediator enables resource provisioning while simultaneously providing migration capability by managing the abstraction and mapping between virtual and physical resources, allowing virtual machines to be moved without direct hardware dependency.
2Ease of manufacture
If virtual machines are tightly bound to underlying hardware resources, then resource allocation is straightforward, but fault tolerance and service continuity deteriorate
Solution Approach 1:
The system segments resource allocation into virtualized units managed by the distributed operating system, allowing resource allocation to remain simple while enabling fault isolation. When physical machines fail, individual virtualized resource segments can be reallocated to different physical hosts without affecting the entire system.
Solution Approach 2:
The distributed operating system dynamically changes the mapping parameters between virtual machines and physical hardware resources. This allows the system to maintain straightforward resource allocation logic while adapting to failures by remapping virtual resources to different physical machines, thereby improving fault tolerance.
3Adaptability or versatility
If live memory of each system is transferred over a network over a long distance for migration, then virtual machine portability is achieved, but migration time and service delays increase
Solution Approach 1:
The patent segments memory resources into virtualized memory units that are managed independently by the distributed operating system. Instead of transferring entire live memory states over the network, the system transfers only the necessary memory mappings and descriptors, dramatically reducing migration time while maintaining portability.
Solution Approach 2:
The distributed operating system uses copying mechanisms to create virtual memory representations that reference physical memory resources without requiring complete memory state transfer. This allows virtual machines to be ported by copying resource descriptors rather than actual memory contents, reducing migration time while achieving portability.
4Reliability
If virtual machines are migrated across geographically distributed data centers, then service continuity is maintained, but network transfer complexity and time increase
Solution Approach 1:
The distributed operating system serves as an intermediary that manages cross-datacenter migrations by abstracting the complexity of network transfers. It handles resource coordination, mapping, and state management across geographically distributed locations, maintaining service continuity while reducing migration complexity through centralized orchestration.
Solution Approach 2:
The distributed operating system implements universal resource management capabilities that work across different geographic locations and data centers. This multi-functional system handles local and remote migrations uniformly, providing service continuity while simplifying complexity through standardized protocols and interfaces.
Data Source
AI summary
A multi-layer architecture is provided for elastic provisioning of virtualized computing resources. The multi-layer architecture comprises a physical hardware layer comprising a plurality of physical computing machines, a distributed operating system layer that aggregates and virtualizes the computing resources, and a virtual machine layer that comprises virtual machines provisioned, by the distributed operating system layer, with virtualized computing resources. Elastic provisioning of virtualized computing resources comprising receiving computing resource information of a plurality of physical computing machines, producing virtualized computing resources by aggregating the received computing resource information of the plurality of physical computing machines, and provisioning the virtualized computing resources among a plurality of virtual machines.