Physical Machine State Capture and Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Server administrators face challenges with physical server sprawl and inefficient management due to the need to manage individual virtual machines, especially in large enterprise environments where hypervisors are inadequate for applications like SAP and Oracle, which require direct physical machine-level execution.
Innovation Solution
The solution involves capturing, storing, and transferring the state of a physical machine using an enhanced Basic Input/Output System (BIOS), allowing for the abstraction and migration of physical machine states, including BIOS, firmware, and operating system states, through a network interface and processor-based apparatus, enabling seamless operation on different physical machines without the need for hypervisors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If hypervisors are used to host multiple virtual machines, then server administrators gain efficiency and portability, but device complexity increases and physical server sprawl occurs
Solution Approach 1:
The patent extracts the machine state (including BIOS, firmware, and operating system state) from the physical machine and stores it externally. This allows the physical machine to be powered off or replaced while preserving its state, eliminating the need for complex hypervisor infrastructure and reducing physical server sprawl.
Solution Approach 2:
The patent creates a copy of the physical machine's operational state (machine profile and machine state) and stores it in a database. This copy can be restored to a different physical machine, enabling portability without requiring hypervisors or complex virtualization infrastructure.
2Reliability
If individual virtual machines are managed in large enterprise environments, then application execution is optimized, but ease of operation deteriorates due to management complexity
Solution Approach 1:
The patent merges the management of multiple virtual machines into a single physical machine by capturing and storing the entire machine state (including all running applications and their states). This allows administrators to manage one physical machine instead of multiple individual virtual machines, significantly simplifying operation while maintaining application reliability.
3Productivity
If physical machines are migrated or replaced, then resource efficiency improves, but loss of time increases due to rebooting requirements
Solution Approach 1:
The patent performs preliminary action by capturing and storing the machine state before the physical machine is powered off or replaced. When the machine needs to be restored, the stored state is reloaded, allowing the machine to resume operation immediately without requiring rebooting, thus eliminating time loss while enabling resource efficiency.
4Reliability
If hypervisors are deployed to provide high availability, then reliability improves, but device complexity and resource consumption increase
Solution Approach 1:
The patent extracts the reliability function from the hypervisor layer and implements it directly at the physical machine level by capturing and storing the complete machine state. This eliminates the need for hypervisor infrastructure while maintaining high availability capabilities, as the stored state can be restored to any physical machine without requiring complex virtualization layers.
Data Source
AI summary
According to one embodiment, an apparatus comprises a physical computing device including a network interface configured to enable communications over a network, and at least one processor. The apparatus captures an operational state of the physical computing device including information pertaining to a state of the physical computing device resulting from operations performed by the physical computing device. The apparatus further configures the physical computing device with the operational state captured from a desired computing device to continue operations of the desired computing device on the physical computing device. Embodiments may further include a method and computer-readable media encoded with software for the storage and transfer of physical machine state in substantially the same manner described above.