Error Event Vector Propagation in Multi-Component Processing Devices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In processing devices with multiple components, effectively propagating error events across the device to enable appropriate responsive actions is challenging, as existing systems lack efficient mechanisms for error event notification and handling.

Innovation Solution

A processing device with a control node that updates an error event vector across components, utilizing event detection circuitry to report events and broadcast updates, allowing each component to perform specific actions in response to error conditions, including notifying event handling software and potentially resetting the device.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If error events are propagated across all components using a centralized notification mechanism, then all components can be informed of error conditions, but the device complexity increases due to the need for centralized control and coordination

Engineering Contradiction:
Improveerror event propagationVSAvoidnotification mechanism
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides the error event notification system into multiple independent error event vector copies, each maintained by different components. Instead of a single centralized error state, each component has its own copy that it can independently update and use, eliminating the need for complex centralized coordination while ensuring all components remain informed of error conditions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates multiple copies of the error event vector across different components. Each component maintains a local copy that mirrors the global error state, allowing components to independently detect and respond to errors without requiring real-time communication or coordination with a central controller.

Inventive Principle:
Principle #26Copying

2Speed

If multiple components independently detect and respond to error events, then response time is reduced, but inconsistency may occur among different components' error states

Engineering Contradiction:
Improveerror response timeVSAvoiderror state consistency
Core Design Contradiction:
SpeedVSStability of the object's composition

Solution Approach 1:

The control node monitors error event vector copies across components and detects inconsistencies. When a component's local copy diverges from the expected state, the control node provides feedback to correct the inconsistency, ensuring all components maintain consistent error state information while allowing independent error detection and response.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The control node acts as an intermediary that coordinates error state consistency across components. It receives error reports from components, updates the master error event vector, and distributes updated copies to relevant components, ensuring synchronization without preventing independent error response actions.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of operation

If a centralized control node manages error event vectors, then coordination is simplified, but the loss of time occurs due to centralized update bottlenecks

Engineering Contradiction:
Improveerror event coordinationVSAvoiderror vector update time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

Components pre-fetch error event vector copies from the control node before they are needed for error detection and response. By having error state information available in advance at each component, the system eliminates update bottlenecks during critical error response operations while maintaining centralized coordination for consistency.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts the update frequency and timing of error event vector copies based on operational conditions. During normal operation, updates occur at optimized intervals, while during error conditions, the system enables rapid propagation of error state information to ensure timely response without unnecessary overhead during stable operation.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS20230281066A1Device Event Notification
Publication Date: 2023.09.07 GRAPHCORE LTD
  • US20230281066A1 patent drawing
  • US20230281066A1 patent drawing
  • US20230281066A1 patent drawing

AI summary

An error event vector is defined for the device, where each element of that error event vector is used to indicate whether or not an event of the associated event class has occurred for any of the components of the device. If so, a control node causes the respective element of each of the copies of the error event vector to be set to indicate that an error of the event class has occurred. A component, i.e. the second one of the components, performs a responsive action for the event class in response to the update to its own copy of the error event vector.