Dynamic Storage Slice Sizing for I/O Workload Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current data storage systems face inefficiencies in managing I/O workloads, as they often use fixed slice sizes for logical address spaces, which can lead to inefficient data movement and resource scalability issues due to varying I/O workloads across different portions of the storage.

Innovation Solution

The method involves dynamically adjusting slice sizes based on current I/O workload densities, allowing for finer granularity in data movement and optimization by partitioning or merging slices according to predetermined criteria, thereby optimizing storage tier utilization and improving performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If fixed slice sizes are used for logical address spaces, then storage system simplicity is maintained, but data movement efficiency deteriorates due to varying I/O workloads

Engineering Contradiction:
Improvedata movement efficiencyVSAvoidslice size management complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies dynamics by transitioning from fixed slice sizes to dynamic slice sizes that automatically adjust based on I/O workload characteristics. The system continuously monitors I/O patterns and reallocates slice sizes to match actual usage, enabling efficient data movement while adapting to changing workload demands without manual intervention.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of slice size from a static value to a dynamic variable that responds to I/O workload conditions. By monitoring I/O metrics and adjusting slice sizes accordingly, the system optimizes data movement efficiency while maintaining manageable complexity through automated parameter adjustment.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If uniform slice sizes are used across all logical address spaces, then system operation simplicity is maintained, but storage resource utilization deteriorates

Engineering Contradiction:
Improvestorage resource utilizationVSAvoidslice management ease
Core Design Contradiction:
ProductivityVSEase of operation

Solution Approach 1:

The patent applies local quality by assigning different slice sizes to different logical address spaces based on their specific I/O workload characteristics. Instead of uniform treatment, the system tailors slice sizes to local usage patterns, optimizing storage resource utilization for each region while maintaining ease of operation through automated detection and adjustment.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system performs self-service by automatically monitoring I/O workloads and adjusting slice sizes without requiring manual administration. The automated mechanisms detect usage patterns and reallocate resources independently, maintaining high storage utilization while preserving operational simplicity.

Inventive Principle:
Principle #25Self-service

3Productivity

If slice sizes are adjusted frequently to match changing workloads, then data storage efficiency is improved, but system stability deteriorates

Engineering Contradiction:
Improvedata storage efficiencyVSAvoidslice configuration stability
Core Design Contradiction:
ProductivityVSStability of the object's composition

Solution Approach 1:

The patent applies periodic action by implementing scheduled or threshold-based adjustments rather than continuous changes. The system monitors I/O workloads and adjusts slice sizes at appropriate intervals or when specific thresholds are crossed, maintaining stability during normal operation while enabling efficiency improvements when conditions warrant change.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system uses feedback mechanisms to monitor I/O workload patterns and trigger slice size adjustments only when necessary. By incorporating feedback loops that detect significant changes in usage patterns, the system maintains stability during normal conditions while adapting to genuine workload changes, balancing efficiency with stability.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS9916090B1Techniques for dynamically adjusting slice size
Publication Date: 2018.03.13 EMC IP HLDG CO LLC
  • US9916090B1 patent drawing
  • US9916090B1 patent drawing
  • US9916090B1 patent drawing

AI summary

Described are techniques for determining slice sizes. First I/O workload information is received for a slice having a corresponding logical address subrange of a logical address range of a logical device. The corresponding logical address subrange is a first size denoting a size of the slice at a first point in time when the slice has a current I/O workload denoted by the first I/O workload information. It is determined, in accordance with the first I/O workload information, whether to adjust the size of the slice. Responsive to determining to adjust the size of the slice, first processing is performed that adjusts the size of the slice such as by partitioning the slice or merging the slice with one or more other adjacent slices.