Availability Domain Aware Replication Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Legacy techniques for implementing replication policies in distributed storage platforms are limited in their ability to be aware of availability domains, leading to replication policy violations and increased hardware and maintenance expenses, especially in highly scalable and dynamic computing systems.
Innovation Solution
The implementation of a heuristics-based informed search technique to efficiently select an availability domain aware replication configuration, which monitors current replication configurations, determines possible configurations, and uses transition scores based on heuristic functions to ensure compliance with availability domain boundaries, reducing memory, processing power, and network bandwidth demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If legacy replication policy techniques are used, then implementation simplicity is maintained, but availability domain awareness is lost leading to replication policy violations
Solution Approach 1:
The system pre-calculates and stores availability domain information for each node, and pre-determines valid replication configurations that satisfy availability domain awareness requirements. When a replication operation is needed, the system queries these pre-computed configurations rather than performing complex real-time calculations, thus ensuring policy compliance without adding operational complexity.
Solution Approach 2:
The patent introduces an intermediary component (replication manager or configurator) that sits between the storage system and the replication policy enforcement mechanism. This intermediary handles the complexity of availability domain awareness by maintaining node-to-availability-domain mappings and automatically selecting valid replication configurations, shielding users from the underlying complexity while ensuring compliance.
2Quantity of substance
If the replication factor is decreased to reduce hardware expenses, then hardware costs are reduced, but data consistency and availability are compromised
Solution Approach 1:
The system applies different replication strategies to different availability domains. Within each availability domain, the system ensures that replicas are distributed across different domains, providing local optimization for fault tolerance. This allows the system to use a lower overall replication factor while maintaining data consistency through intelligent placement of replicas across availability domains.
3Reliability
If more availability domains are added to reduce replication policy violations, then replication compliance is improved, but hardware and facility expenses increase significantly
Solution Approach 1:
The system implements partial availability domain awareness by focusing on the critical aspect of ensuring at least one replica resides in a different availability domain, rather than requiring complete distribution across all possible domains. This partial implementation achieves sufficient replication compliance without the need to expand the number of availability domains or add excessive hardware resources.
4Ease of operation
If random node selection is used for replication, then implementation simplicity is maintained, but availability domain awareness is lost
Solution Approach 1:
The system changes the selection parameters from purely random to a constrained random selection that incorporates availability domain information. The replication manager queries the availability domain of each candidate node and filters out nodes that would violate the availability domain awareness policy, then performs random selection within the valid set. This maintains operational simplicity while ensuring compliance.
Data Source
AI summary
Systems for distributed data storage. A method commences upon accessing a set of data items that describe computing nodes to be organized into a ring topology. The ring topology and distributed data storage policies are characterized by quantitative failure-resilient characteristics such as a replication factor. Various characteristics of the topology serve to bound two or more availability domains of the ring into which the computing nodes can be mapped. A set of quantitative values pertaining to respective quantitative failure-resilient characteristics are used for enumerating candidate ring topologies where the computing nodes are mapped into the availability domains. Using the quantitative failure-resilient characteristics, alternative candidate ring topologies are evaluated so as to determine a configuration score for candidate ring topologies. A candidate ring topology is configured based on a computed configuration score surpassing a threshold score. When a failure event is detected, the ring is reevaluated, remapped, and considered for reconfiguration.