Database Storage Controller Block-Level QoS Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Storage systems struggle to differentiate between various types of I/O operations, leading to inefficiencies in quality of service policies that often benefit only a fraction of storage requests, resulting in wastage of resources.

Innovation Solution

Implementing a database classifier for each I/O request using POSIX commands, allowing for classification into specific database classes, and associating these classes with tailored quality of service policies stored in a table, enabling the storage controller to enforce policies on a block-level basis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If quality of service policies are implemented at a volume-wide level, then storage system simplicity is maintained, but resource efficiency deteriorates as policies benefit only a fraction of storage requests

Engineering Contradiction:

Improvestorage system simplicityVSAvoidresource efficiency

Core Design Contradiction:

Device complexityVSLoss of energy

Solution Approach 1:

The patent segments the storage system into two hierarchical levels: volume-wide quality of service policies and block-level quality of service policies. This segmentation allows the system to maintain simplicity at the volume level while introducing fine-grained control at the block level, thereby improving resource efficiency without completely sacrificing system simplicity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by allowing different quality of service policies to be applied to different blocks within the same volume. Each block can have its own quality of service policy based on its specific requirements, enabling precise resource allocation and eliminating the waste associated with applying uniform volume-wide policies to all blocks.

Inventive Principle:

Principle #3Local quality

2Adaptability or versatility

If block interface abstraction is used, then interoperability between database systems and storage systems is improved, but information about I/O operation types is lost

Engineering Contradiction:

ImproveinteroperabilityVSAvoidI/O operation type information

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent introduces a block interface layer as an intermediary between the database system and the storage system. This intermediary maintains interoperability by providing a standardized interface while simultaneously capturing and transmitting I/O operation type information through classification mechanisms, thus preventing information loss despite the abstraction.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent performs preliminary classification of I/O operations at the block interface level before they are processed by the storage system. By classifying I/O operations into different types (e.g., random read, sequential write, metadata access) in advance, the system preserves operation type information that can then be used to apply appropriate quality of service policies without compromising interoperability.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9842156B2System and method for providing differentiated storage service in a database

Publication Date: 2017.12.12 INTEL CORP

AI summary

In accordance with some embodiments, classification of input/output requests from a database to a storage system may be performed. Each input/output request may be associated with a database class, and each database class may be mapped to a quality of service policy. Thus, quality of service may be enforced such that different data blocks within the storage system of the database may be afforded appropriate quality of service.