PCIe Failure Location Mapping in Composable Server Infrastructure

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In server systems with composable disaggregation infrastructure, identifying a failed PCIe card can be time-consuming and challenging, especially when the number and positions of PCIe boxes dynamically change, leading to difficulties in recognizing the BDF of each PCIe card and its position, which impairs system availability and requires skilled user intervention.

Innovation Solution

A management apparatus that stores management information associating device identification with positional information, enabling the processor to quickly identify the position of a failed device by correlating received BDF information with pre-stored box and device information tables, facilitating rapid notification of the device's location to the user.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If PCIe boxes are dynamically added or removed from the server system, then system adaptability and flexibility are improved, but the ability to identify and locate PCIe cards deteriorates

Engineering Contradiction:
Improvesystem flexibilityVSAvoidPCIe card identification difficulty
Core Design Contradiction:
Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The management apparatus pre-stores correspondence tables mapping PCIe card BDFs to their physical positions in PCIe slots, and PCIe slot positions to PCIe box locations, before any failure occurs. This preliminary preparation enables immediate identification of failed cards without requiring system reconfiguration or manual searching, thus resolving the contradiction between dynamic adaptability and identification difficulty

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The management apparatus acts as an intermediary between the BMC (which detects failures) and the user (who needs to locate and replace failed cards). It maintains and queries correspondence tables to translate BDF information into physical location information, bridging the gap between logical device identification and physical location without requiring direct user intervention in the complex dynamic system

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the BMC converts BDF to slot number using correspondence tables, then failure identification accuracy is improved, but maintenance time increases due to manual intervention requirements

Engineering Contradiction:
Improvefailure identification accuracyVSAvoidmaintenance time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The management apparatus provides automated feedback to the user by sending notifications that include the specific PCIe box location and slot number where the failed card is installed. This feedback mechanism eliminates the need for users to manually interpret BDF information or search through correspondence tables, maintaining high identification accuracy while dramatically reducing maintenance time

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system enables self-service maintenance by automatically generating and sending failure location information to users via notification messages. Users receive ready-to-act information including which PCIe box to open and which slot to check, eliminating the need for skilled manual intervention and reducing maintenance time while preserving accurate failure identification

Inventive Principle:
Principle #25Self-service

3Ease of operation

If the system maintains detailed correspondence tables for all PCIe cards, then device location identification capability is improved, but system complexity increases

Engineering Contradiction:
Improvedevice location identification easeVSAvoidmanagement information complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The management apparatus segments the correspondence information into two separate tables: one mapping BDFs to PCIe slot numbers, and another mapping PCIe slot numbers to PCIe box locations. This segmentation reduces the complexity of any single data structure while maintaining the complete tracking capability, making the system easier to manage and query without sacrificing identification ease

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS12547486B2Management apparatus, information processing system, and management method
Publication Date: 2026.02.10 FSAS TECH INC
  • US12547486B2 patent drawing
  • US12547486B2 patent drawing
  • US12547486B2 patent drawing

AI summary

A memory stores management information where identification information of each of a plurality of devices used by an information processing apparatus, first positional information, and second positional information are associated with one another. The first positional information indicates a position of a device storage storing the plurality of devices. The second positional information indicates a storage position of each of the plurality of devices in the device storage. A processor receives failed device information including identification information of a failed device among the plurality of devices from the information processing apparatus. The processor identifies the position of the device storage storing the failed device and the storage position of the failed device in the device storage from the identification information of the failed device included in the failed device information on the basis of management information.