LRU Queue Compression for Buffer Cache Miss Reduction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional data storage systems face performance issues due to cache misses when data aged off the LRU queue is re-accessed shortly thereafter, leading to costly disk accesses, as the cache capacity is not effectively utilized.

Innovation Solution

The method involves compressing cache pages from the LRU queue to extend their retention in the cache, allowing data to remain available upon subsequent re-access by rearranging the LRU queue structure and compressing buffer cache pages at the head end, thereby freeing memory for reuse.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If data is placed on LRU queue to manage cache space, then cache capacity is maintained, but data may be re-accessed shortly after aging off resulting in cache misses and disk accesses

Engineering Contradiction:

Improvecache capacityVSAvoiddata availability

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The LRU queue is segmented into two distinct portions: a compressed portion at the head end containing compressed buffer cache pages, and an uncompressed portion at the tail end containing regular buffer cache pages. This segmentation allows different retention strategies for different data segments, preventing premature eviction of frequently accessed data while maintaining cache capacity management.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter of data retention by compressing buffer cache pages in the compressed portion of the LRU queue. Compression extends the retention time of data, allowing it to remain in cache longer and be available for subsequent re-accesses, thereby reducing cache misses and disk accesses.

Inventive Principle:

Principle #35Parameter changes

2Duration of action of moving object

If more entries are added to LRU queue to extend useful capacity, then data retention is improved, but cache memory space is consumed

Engineering Contradiction:

Improvedata retention timeVSAvoidcache memory space

Core Design Contradiction:

Duration of action of moving objectVSQuantity of substance

Solution Approach 1:

The patent applies compression to buffer cache pages in the compressed portion of the LRU queue, changing their physical state to reduce the memory space they occupy. This allows more data entries to be retained in the cache with the same memory capacity, extending useful cache capacity without requiring additional memory allocation.

Inventive Principle:

Principle #35Parameter changes

3Quantity of substance

If buffer cache pages are compressed to free memory, then cache space utilization is improved, but system complexity increases

Engineering Contradiction:

Improvecache space utilizationVSAvoidLRU queue structure

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The LRU queue is divided into compressed and uncompressed portions, with compressed buffer cache pages stored in a compressed portion at the head end and uncompressed pages in the tail end. This segmentation simplifies the compression management by confining compressed data to a specific region, making it easier to track and manage compared to compressing all cache pages uniformly.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS9727479B1Compressing portions of a buffer cache using an LRU queue

Publication Date: 2017.08.08 EMC IP HLDG CO LLC

AI summary

Techniques are described for compressing cache pages from an LRU (Least-Recently-Used) queue so that data takes longer to age off and be removed from the cache. This increases the likelihood that data will be available within the cache upon subsequent re-access, reducing the need for costly disk accesses due to cache misses.