Demote Threads Cache Management

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current cache management systems face challenges in efficiently demoting tracks from the cache to free up segments, leading to potential cache depletion and queuing of I/O requests due to an imbalance between the rate of cache segment consumption and demotion.

Innovation Solution

Implementing multiple demote ready lists across processors, allowing demote threads and I/O threads to concurrently process demotion tasks, with a demote scheduler determining the number of demote threads based on free cache segment thresholds to maintain adequate cache space.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a single demote thread processes the demote ready list sequentially, then the cache management system maintains simplicity in thread management, but the rate of track demotion is insufficient leading to cache depletion and I/O request queuing

Engineering Contradiction:
Improvetrack demotion rateVSAvoidthread management complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent divides the single demote ready list into multiple separate demote ready lists, with each list processed by a dedicated demote thread. This segmentation allows parallel processing of track demotion operations across multiple threads simultaneously, significantly increasing the overall demotion rate while maintaining manageable complexity through structured list organization

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent combines multiple demote threads operating on separate demote ready lists into a coordinated system where all threads work together to process the complete set of demote-ready tracks. The demote scheduler merges the progress of multiple threads to ensure comprehensive cache management, achieving high productivity through parallel operation while maintaining system coherence

Inventive Principle:
Principle #5Merging (Combining)

2Productivity

If multiple demote threads concurrently process multiple demote ready lists, then the track demotion rate increases significantly, but the thread management complexity increases

Engineering Contradiction:
Improvetrack demotion rateVSAvoidthread coordination complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The demote scheduler implements feedback mechanisms that continuously monitor the state of multiple demote ready lists and adjust thread execution accordingly. The scheduler receives feedback about which lists contain demote-ready tracks and dynamically directs threads to process the appropriate lists, ensuring efficient resource utilization and maintaining system coherence despite parallel operation

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The demote scheduler acts as an intermediary between multiple demote threads and the cache management system. It coordinates thread execution, manages the distribution of demote ready lists to threads, and ensures that demotion operations proceed in a controlled manner, thereby reducing the effective complexity of managing multiple concurrent threads

Inventive Principle:
Principle #24Intermediary (Mediator)

3Quantity of substance

If the cache segments are consumed faster than tracks are demoted, then the cache can store more recently accessed tracks, but the free cache segments deplete leading to I/O request queuing

Engineering Contradiction:
Improvecache storage capacityVSAvoidI/O request latency
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The system performs preliminary actions by proactively identifying and moving tracks to the demote ready list before the cache becomes completely full. The demote scan continuously monitors cache state and prepares tracks for demotion in advance, ensuring that when I/O requests arrive, sufficient cache segments are available, thereby preventing queuing and reducing latency

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent establishes continuous demotion action through multiple concurrently running demote threads that constantly process demote ready lists. This continuous demotion activity ensures a steady supply of freed cache segments, matching or exceeding the rate at which segments are consumed by I/O operations, thereby maintaining adequate cache capacity and preventing request queuing

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentUS9971689B2Invoking input/output (I/O) threads and demote threads on processors to demote tracks from a cache
Publication Date: 2018.05.15 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US9971689B2 patent drawing
  • US9971689B2 patent drawing
  • US9971689B2 patent drawing

AI summary

Provided are a computer program product, system, and method for invoking Input/Output (I/O) threads and demote threads on processors to demote tracks from a cache. An Input/Output (I/O) thread, executed by a processor, processes I/O requests directed to tracks from the storage stored in the cache. A demote thread, executed by the processor, processes a demote ready list, indicating tracks eligible to demote from cache, to select tracks to demote from the cache to free cache segments in the cache. After processing a number of I/O requests, the I/O thread processes the demote ready list to demote tracks from the cache in response to determining that a number of free cache segments in the cache is below a free cache segment threshold.