Selective Space Reclamation in Data Storage Memory Using Heat and Relocation Metrics

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current data storage systems face inefficiencies in reclaiming memory space, particularly in secondary cache memories where data relocation incurs a high number of I/O operations due to the log-structured nature of data storage, leading to resource wastage and performance issues.

Innovation Solution

The method involves determining heat and relocation metrics for data stored in the memory, calculating utility metrics based on these, and invalidating data that fails a utility metric threshold to avoid unnecessary relocation, thereby prioritizing the removal of less frequently accessed data and exempting recently added or policy-designated data from reclamation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If data is relocated in log-structured extents to reclaim space, then space reclamation is achieved, but the number of I/O relocation operations increases significantly

Engineering Contradiction:

Improvefree space in memoryVSAvoidI/O relocation operations

Core Design Contradiction:

Quantity of substanceVSProductivity

Solution Approach 1:

The patent changes the parameter being optimized from pure space reclamation to a composite utility metric that balances space reclamation with I/O operation minimization. By calculating utility metrics that incorporate both heat metrics (data value) and relocation metrics (I/O cost), the system selects data for invalidation based on this composite parameter, thereby achieving space reclamation while controlling the number of relocation operations.

Inventive Principle:

Principle #35Parameter changes

2Quantity of substance

If all invalidated data is relocated, then space is fully reclaimed, but resource wastage occurs due to moving valuable data

Engineering Contradiction:

Improvereclaimed spaceVSAvoidresource wastage

Core Design Contradiction:

Quantity of substanceVSLoss of energy

Solution Approach 1:

The patent applies local quality by differentiating between different data based on their heat metrics and utility metrics. Instead of treating all invalidated data uniformly, the system identifies and preserves high-value data (those with high utility metrics) while invalidating low-value data. This selective approach ensures that space reclamation does not result in the wasteful relocation of valuable data that should be retained.

Inventive Principle:

Principle #3Local quality

3Speed

If data is kept in cache memory, then access speed is improved, but storage capacity is reduced

Engineering Contradiction:

Improvedata access speedVSAvoidavailable storage capacity

Core Design Contradiction:

SpeedVSVolume of stationary object

Solution Approach 1:

The patent implements a dynamic space reclamation strategy that adapts to changing system conditions. The utility metric threshold is dynamically adjusted based on system state, allowing the system to optimize the balance between retaining data for fast access and making space available. This dynamic approach ensures that the cache memory maintains optimal performance while maximizing available storage capacity when needed.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS10482032B2Selective space reclamation of data storage memory employing heat and relocation metrics

Publication Date: 2019.11.19 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Space of a data storage memory of a data storage memory system is reclaimed by determining heat metrics of data stored in the data storage memory; determining relocation metrics related to relocation of the data within the data storage memory; determining utility metrics of the data relating the heat metrics to the relocation metrics for the data; and making the data whose utility metric fails a utility metric threshold, available for space reclamation.