NVM Controller Logic for Remote Failure Isolation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current systems for remote data retrieval and failure isolation in complex computer servers are costly and inefficient, requiring on-site visits or replacing entire systems due to the lack of precise diagnostic data from remote servers.

Innovation Solution

Implementing Non-Volatile Memory (NVM) devices, such as flash memory and STTRAM, with NVM controller logic to store and retrieve data persistently, allowing for remote access and analysis of debug data, enabling precise failure isolation and reducing support costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If field engineers are sent to remote servers or entire systems are replaced to resolve issues, then reliability and serviceability are improved, but support costs and time consumption increase significantly

Engineering Contradiction:

Improveserver reliabilityVSAvoidtime for on-site visits and system replacement

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by continuously collecting and storing debug data, telemetry information, and system state data in non-volatile memory before failures occur. This pre-collection of diagnostic information enables immediate remote analysis when issues arise, eliminating the need for time-consuming on-site visits and rapid system replacement while maintaining high reliability.

Inventive Principle:

Principle #10Preliminary action

2Ease of operation

If debug data is collected and stored for remote analysis, then ease of operation and cost efficiency are improved, but device complexity increases due to additional firmware agents and microcontrollers

Engineering Contradiction:

Improveremote data retrieval easeVSAvoidfirmware agent complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent applies merging by integrating multiple data collection functions (telemetry, debug data, system state monitoring) into a unified non-volatile memory storage system. This consolidation reduces the need for separate complex firmware agents and independent microcontrollers, simplifying the overall device architecture while maintaining ease of remote operation and data retrieval.

Inventive Principle:

Principle #5Merging (Combining)

3Reliability

If entire systems are replaced instead of individual defective units, then reliability is ensured, but loss of substance and cost increase due to replacing functional components

Engineering Contradiction:

Improvesystem reliabilityVSAvoidfunctional components replaced

Core Design Contradiction:

ReliabilityVSLoss of substance

Solution Approach 1:

The patent implements segmentation by dividing the system into independently diagnosable units through precise failure isolation using collected debug data. This enables identification and replacement of only the specific defective component rather than the entire system, maintaining reliability while preserving functional components and reducing material waste.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS9912474B2Performing telemetry, data gathering, and failure isolation using non-volatile memory

Publication Date: 2018.03.06 SK HYNIX NAND PRODUCT SOLUTIONS CORP

AI summary

Methods and apparatus related to performance of telemetry, data gathering, and failure isolation using non-volatile memory are described. In one embodiment, a Non-Volatile Memory (NVM) controller logic stores data in a portion of an NVM device. The portion of the NVM device is determined based at least in part on a type or an identity of a sender of the data. Also, the data is encrypted in accordance with a public key provided by the sender. Other embodiments are also disclosed and claimed.