Command Stream Data Save Indicators for Task Suspension

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data processing systems lack efficient methods for suspending and resuming processing tasks, particularly in scenarios where data created or changed during task execution needs to be preserved for later resumption.

Innovation Solution

The implementation of data save indicators within command streams allows for the identification and saving of data that needs to be preserved upon suspension of processing tasks, eliminating the need for active tracking using markers or 'dirty bits'.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is saved to main memory upon suspension of processing tasks, then data loss is prevented and task resumption is enabled, but memory access time increases and system productivity decreases

Engineering Contradiction:
Improvedata preservationVSAvoidsystem efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by pre-identifying and tagging data that requires saving through data save indicators embedded in the command stream. This allows the processing resource to know in advance which data must be preserved, enabling faster suspension operations without ad-hoc analysis of which data to save.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The invention extracts only the specific data that needs to be saved from the working memory, rather than saving all data. By using data save indicators to identify and extract only the necessary data portions, the system reduces the amount of memory I/O operations required during suspension and resumption, thereby improving productivity while maintaining reliability.

Inventive Principle:
Principle #2Taking out (Extraction)

2Reliability

If all data in working memory is saved upon suspension, then complete task state is preserved for accurate resumption, but memory access time and energy consumption increase

Engineering Contradiction:
Improvetask state preservationVSAvoidenergy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system extracts and saves only the specific data portions that are marked by data save indicators, rather than saving the entire working memory contents. This selective extraction approach preserves the necessary task state for accurate resumption while significantly reducing the energy consumption associated with memory write operations.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Different portions of working memory are treated differently based on their importance to task resumption. Data save indicators create a local quality distinction, marking specific data regions as requiring preservation while leaving other data to be discarded. This enables energy-efficient selective saving of only the critical task state information.

Inventive Principle:
Principle #3Local quality

3Measurement precision

If markers or dirty bits are used to track data changes, then data tracking precision is improved, but device complexity increases

Engineering Contradiction:
Improvedata tracking accuracyVSAvoidtracking mechanism complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The invention removes the complex marker and dirty bit tracking mechanisms from the system. Instead of using additional hardware or software structures to track data changes, the system extracts the tracking function by embedding simple data save indicators directly in the command stream, achieving the same data identification purpose with reduced complexity.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Data save indicators serve as an intermediary mechanism that replaces complex marker systems. These indicators are embedded in the existing command stream structure and provide the necessary data identification functionality without requiring separate tracking infrastructure, thus reducing device complexity while maintaining data tracking accuracy.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS20250138897A1Data processing systems
Publication Date: 2025.05.01 ARM LTD
  • US20250138897A1 patent drawing
  • US20250138897A1 patent drawing
  • US20250138897A1 patent drawing

AI summary

In a data processing system, a command stream provided to a processing resource to cause the processing resource to perform a processing task for an application executing on a host processor comprises a sequence of commands for execution by the processing resource to cause the processing resource to perform the processing operations for the processing task and one or more data save indicators that indicate data that is to be saved. In response to the processing resource receiving a request to suspend processing of the processing task, data indicated by one of the one or more data save indicators in the command stream is stored in memory.