Persistent Memory Data Protection via Chassis Power Pooling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information handling systems with persistent memory face data loss during partial power loss due to insufficient power capacity in modular chassis environments, where a failed local sled power source can lead to a failed persistent memory save operation, resulting in chassis shutdown and data loss for all nodes.
Innovation Solution
A method is implemented in a host information handling system to determine the runtime health status of persistent memory subsystems and communicate this status to a management module, which then powers on energy storage devices and keeps healthy power supply units on to ensure all systems complete their save operations before shutting down, thereby preventing data loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If local sled power sources are sized to support operation during persistent memory save operations, then data protection is improved, but power capacity requirements and system complexity increase
Solution Approach 1:
The patent combines multiple local sled power sources into a shared power pool at the chassis level. Instead of each sled having independently sized power sources, the power sources are merged and current sharing is enabled, allowing the collective power capacity to support persistent memory save operations across multiple sleds simultaneously.
Solution Approach 2:
The chassis-level power management system provides multi-functionality by serving both individual sled power requirements and collective persistent memory save operations. The power management module can dynamically allocate power from the shared pool to multiple sleds during save operations, making the power infrastructure universally applicable to both individual and system-wide needs.
2Use of energy by moving object
If current sharing is enabled between power sources, then power utilization efficiency is improved, but control complexity and risk of cascading failures increase
Solution Approach 1:
The power management module implements feedback control by continuously monitoring the status of power sources and sleds, then dynamically adjusting power allocation. When a sled completes its save operation or a power source status changes, the system receives feedback and reallocates power accordingly, optimizing power utilization while maintaining system stability.
Solution Approach 2:
The power management module acts as an intermediary between individual sled power sources and the chassis power distribution system. It mediates the current sharing process, managing the complexity of coordinating multiple power sources and preventing cascading failures by controlling power flow at the chassis level rather than allowing uncoordinated sharing.
3Reliability
If a single local sled power source fails during persistent memory save operation, then system resilience is tested, but data loss risk increases due to insufficient remaining power capacity
Solution Approach 1:
The system implements beforehand cushioning by maintaining a reserved power capacity in the shared power pool that can compensate for individual power source failures. The power management module is configured to detect power source failures and reallocate the reserved power capacity to support ongoing persistent memory save operations on other sleds, preventing data loss despite the failure.
4Ease of operation
If chassis infrastructure is powered during save operations from common system voltage rail, then operational continuity is improved, but power consumption and heat generation increase
Solution Approach 1:
The system applies dynamics by making the power supply configuration adaptive rather than static. During persistent memory save operations, the power management module dynamically adjusts the power source configuration, switching certain infrastructure components to be powered from the shared power pool while maintaining operational continuity. This dynamic reconfiguration optimizes power consumption based on the operational state of the system.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach ensures that data is preserved by allowing healthy power supply units to maintain power until all systems complete their save operations, reducing the risk of data loss during partial power loss events in modular chassis environments.
Implementation Method 1
A battery, capacitor, or other energy storage device either internal or external to the NVDIMM may supply electrical energy for a 'save' operation
Implementation Method 2
A battery, capacitor, or other energy storage device either internal or external to the NVDIMM may supply electrical energy
Implementation Method 3
certain portions of the chassis infrastructure may be required to be powered from the chassis common system voltage rail
Data Source
AI summary
A method may include, in a host information handling system configured to be inserted into a chassis providing a common hardware infrastructure to a plurality of modular information handling systems including the information handling system: (i) determining a runtime health status of a persistent memory subsystem of the host information handling system; and (ii) communicating a health status indicator indicative of the runtime health status to a management module configured to manage the common hardware infrastructure.