Write Cache Buffer Sizing via Runtime Energy Parameters

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional storage processors face limitations in determining the optimal size of the write cache during emergency data save procedures due to reliance on worst-case assumptions, which can result in reduced performance and inefficient energy usage.

Innovation Solution

The method involves dynamically retrieving operating parameters such as energy storage capacity, data transfer rate, and energy consumption rate at runtime to determine the maximum size of the write cache, allowing for instance-specific customization and efficient allocation from volatile memory.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If worst-case assumptions are used to determine write cache size, then reliability is improved, but productivity deteriorates

Engineering Contradiction:

Improveemergency data save reliabilityVSAvoiddata transfer rate

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent changes the parameter determination approach from fixed worst-case assumptions to dynamic runtime retrieval of actual operating parameters (energy storage capacity, data transfer rate, energy consumption rate). This allows the write cache size to be optimized based on actual hardware capabilities rather than conservative estimates, resolving the contradiction between reliability and productivity.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If larger write cache is allocated, then productivity is improved, but use of energy worsens

Engineering Contradiction:

Improvewrite cache capacityVSAvoidenergy consumption during power-loss save

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent implements feedback by retrieving actual operating parameters at runtime (energy consumption rate of transfer logic, data transfer rate, energy storage capacity) and using these values to dynamically determine the optimal write cache size. This feedback mechanism ensures energy-efficient cache allocation matched to actual hardware characteristics rather than using fixed conservative estimates.

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If instance-specific parameters are retrieved at runtime, then adaptability is improved, but device complexity worsens

Engineering Contradiction:

Improvehardware configuration adaptabilityVSAvoidparameter retrieval and calculation complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by having the system retrieve and store operating parameters (energy consumption rate, data transfer rate, energy storage capacity) during initialization or setup phases. These parameters are then available for runtime use without requiring complex real-time calculations, thus achieving adaptability while minimizing runtime complexity.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9767030B2Sizing a write cache buffer based on emergency data save parameters

Publication Date: 2017.09.19 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Embodiments relate to saving data upon loss of power. An aspect includes sizing a write cache buffer based on parameters related to carrying out this emergency data save procedure. A computer implemented method for allocating a write cache on a storage controller includes retrieving, at run-time by a processor, one or more operating parameters of a component used in a power-loss save of the write cache. The component is selected from the group consisting of an energy storage element, a non-volatile memory, and a transfer logic. A size for the write cache on the storage controller is determined, based on the one or more operating parameters. A write cache, of the determined size, is allocated from a volatile memory coupled to the storage controller.