Site Cache Eviction for Distributed File System Reliability

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Distributed file systems face performance degradation due to geographically diverse data storage, leading to slower access for clients far from data centers, and site caches are prone to hardware failures resulting in data loss.

Innovation Solution

Implementing a site cache at client locations with high bandwidth, low latency connections to process read and write requests, and employing a metadata server to manage data access, along with mechanisms for data replication and eviction policies to prevent data loss from hardware failures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is stored geographically diverse across multiple data centers, then system reliability and fault tolerance are improved, but data access speed and performance deteriorate for clients far from data centers

Engineering Contradiction:

Improvesystem reliabilityVSAvoiddata access speed

Core Design Contradiction:

ReliabilityVSSpeed

Solution Approach 1:

The patent segments the distributed file system into multiple geographically distributed data centers, each capable of independently storing and serving data. This allows clients to access data from the nearest data center, improving access speed while maintaining system reliability through geographic distribution.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements data replication across multiple data centers, creating copies of the same data in different geographic locations. This enables clients to access data from the closest data center copy, reducing latency while preserving data availability and reliability through multiple replicas.

Inventive Principle:

Principle #26Copying

2Speed

If site cache capacity is increased to improve data access performance, then data access speed is improved, but hardware failure risk and potential data loss increase

Engineering Contradiction:

Improvedata access speedVSAvoiddata loss risk

Core Design Contradiction:

SpeedVSReliability

Solution Approach 1:

The patent implements preliminary actions by continuously monitoring site cache health metrics and proactively detecting hardware failure indicators before actual failures occur. This allows the system to take preventive measures such as data evacuation or cache invalidation, maintaining fast local access while preventing data loss from undetected hardware failures.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent establishes feedback mechanisms that continuously monitor site cache performance and health status, using this information to dynamically adjust cache capacity and data placement decisions. This feedback loop enables the system to optimize data access speed while maintaining reliability by responding to changing cache conditions.

Inventive Principle:

Principle #23Feedback

3Productivity

If site cache is used for local data access, then data access performance is improved, but data loss from hardware failure occurs without proper protection mechanisms

Engineering Contradiction:

Improvedata access performanceVSAvoiddata loss

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent applies beforehand cushioning by implementing redundant data storage and protection mechanisms at site caches before hardware failures occur. This includes maintaining data replicas, implementing error correction codes, and establishing recovery procedures in advance, allowing the system to tolerate hardware failures without data loss while maintaining fast local access performance.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentUS10452619B1Decreasing a site cache capacity in a distributed file system

Publication Date: 2019.10.22 EMC IP HLDG CO LLC

AI summary

An indication to retire a computer readable storage medium from the site cache is received. A plurality of data objects stored on the computer readable storage medium are replicated to at least one target computer readable storage medium in the site cache. The computer readable storage medium is removed from the site cache.