Load-Balanced DSN Shutdown Handling for Continuous Data Access

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current dispersed storage networks face challenges in efficiently managing storage unit shutdowns and data integrity across geographically distributed systems, particularly in maintaining data availability and security without redundant copies, and in balancing load across computing devices.

Innovation Solution

A dispersed storage network architecture that includes a managing unit for establishing error encoding parameters, an integrity processing unit for rebuilding missing data slices, and a load-balancing module to select available computing devices for data access, utilizing Cauchy Reed-Solomon encoding for data dispersal and error correction, and a DSN interface to mimic conventional file systems for seamless data storage and retrieval.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If storage units are shut down for maintenance or updates, then device availability is reduced, but data security and integrity must be maintained

Engineering Contradiction:
Improvedata integrityVSAvoidstorage availability
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent segments data into multiple dispersed slices distributed across different storage units. This allows the system to maintain data integrity even when individual storage units are shut down for maintenance, as data can be reconstructed from remaining slices. The load-balancer redistributes access loads to available storage units during shutdown periods.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The load-balancer proactively monitors storage unit status and preemptively redirects access requests before shutdowns occur. It maintains an updated map of storage unit availability and pre-distributes access loads to ensure seamless transitions during maintenance windows without compromising data integrity.

Inventive Principle:
Principle #10Preliminary action

2Quantity of substance

If data is dispersed across multiple storage units without redundant copies, then storage efficiency is improved, but data availability during unit shutdown becomes challenging

Engineering Contradiction:
Improvestorage efficiencyVSAvoiddata availability
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The load-balancer dynamically adjusts data access routing based on real-time storage unit availability. When storage units are online, data is accessed from the optimal location. When units are shut down, the load-balancer dynamically redirects requests to alternative storage units that hold the necessary data slices, maintaining availability without requiring redundant copies.

Inventive Principle:
Principle #15Dynamics

3Productivity

If access requests are distributed across all storage units, then load balancing is achieved, but performance degrades when some units are unavailable

Engineering Contradiction:
Improveload distributionVSAvoiddata access speed
Core Design Contradiction:
ProductivityVSSpeed

Solution Approach 1:

The load-balancer maintains real-time awareness of storage unit status and preemptively routes access requests only to available units. This prevents the performance degradation that would occur if requests were sent to unavailable units, while still achieving load distribution across the available storage infrastructure.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The load-balancer continuously monitors storage unit availability and uses this feedback to dynamically adjust request routing. When units become unavailable, the feedback mechanism triggers immediate rerouting of their workload to remaining available units, maintaining optimal access speed while preserving load balancing principles.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11543963B1Storage unit shutdown in a distributed storage network using a load-balancer
Publication Date: 2023.01.03 PURE STORAGE INC
  • US11543963B1 patent drawing
  • US11543963B1 patent drawing
  • US11543963B1 patent drawing

AI summary

A method begins by a load balancing module of a distributed storage network (DSN) determining availability of a plurality of DSN processing units of a set of DSN processing units based on availability information associated with the plurality of DSN processing units and in response to determined availability, selecting a DSN processing unit form the set to process a data access request. The method continues with the load balancing module receiving an indication that the DSN processing unit is no longer available from the DSN processing unit while the DSN processing unit continues to process previously pending data access requests. The method continues with the load balancing module cancelling selection of the DSN processing unit to process the data access request; and receiving a second indication from the DSN processing unit indication that the DSN processing unit is available.