Storage Volume Swapping via Hot Spot Service Time Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional data storage system optimization techniques fail to accurately distinguish between devices with moderate but steady activity and those with sporadic high activity, leading to suboptimal performance improvements due to reliance on average access activity data across fixed time windows or entire measurement intervals.

Innovation Solution

A method that collects workload statistics in sampling intervals to calculate service times, identifies the highest service times (hot spots), and uses these to rank devices for logical volume swapping, focusing on periods of peak activity to improve system performance by reducing service times without degrading less busy devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If average access activity data over entire measurement intervals is used for volume swapping decisions, then all access activity data is utilized, but devices with sporadic high activity cannot be distinguished from devices with moderate steady activity

Engineering Contradiction:

Improveamount of access activity data utilizedVSAvoidability to distinguish activity patterns

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent extracts only the relevant portion of access activity data by identifying and selecting the highest service time values (hot spots) from the measurement interval, rather than using all data points. This extraction allows the system to focus on peak activity periods while ignoring moderate steady activity, thereby resolving the contradiction between utilizing all data and achieving precise measurement of problematic devices.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies local quality by treating different portions of the access activity data differently. Instead of uniform treatment of all data points, the system identifies specific high-impact periods (hot spots) and gives them disproportionate weight in the swapping decision process. This selective focus on critical periods enables precise identification of devices needing optimization without being diluted by moderate activity data.

Inventive Principle:

Principle #3Local quality

2Measurement precision

If fixed performance time windows are used to restrict analysis periods, then analysis focuses on critical periods, but devices with variable timing of high activity cannot be captured

Engineering Contradiction:

Improvefocus on critical performance periodsVSAvoidability to capture variable timing patterns

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamics by making the selection of analysis periods adaptive rather than fixed. The system dynamically identifies hot spots based on actual service time measurements, allowing the analysis window to shift and expand to capture peak activity periods regardless of when they occur. This dynamic approach enables the system to adapt to variable timing patterns while maintaining focus on critical performance periods.

Inventive Principle:

Principle #15Dynamics

3Ease of operation

If volume swapping is performed based on average service times, then overall system balance is improved, but peak performance during high activity periods is not optimized

Engineering Contradiction:

Improveoverall system balanceVSAvoidpeak performance during high activity

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The patent applies parameter changes by shifting the basis for swapping decisions from average service times to hot spot service times. This parameter change transforms the optimization objective from achieving overall system balance to maximizing peak performance during high-activity periods. The system maintains ease of operation by using the same volume swapping mechanism, but changes the metric used to identify which volumes should be swapped.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS7363453B1Method and apparatus for improving storage device performance by selective volume swapping based on hot spot analysis

Publication Date: 2008.04.22 EMC IP HLDG CO LLC

AI summary

An optimizer process in a storage system automatically selects access activity data for storage devices in the system during periods of interest so that a volume-swapping optimization analysis is based on desired device performance information and thus yields improved optimization results. For each of a number of sampling intervals in an analysis period, workload statistics are collected for logical volumes in the system. From the workload statistics, a service time is calculated for each storage device for each sampling interval. A set of highest service times is identified from the service times for each device, such as those above the 80th percentile, and a service time measure is generated from the set of highest service times. Devices having the highest and lowest measures are identified, and a pair of volumes is identified that, if swapped between the two devices, would improve the service time measure of the higher-service-time device without unduly degrading the service time measure of the lower-service-time device. Such volumes are subsequently swapped, improving the performance of the higher-service-time device and the overall system.