Any-Local-Quorum Reads in Multi-Site Database Replication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed databases face inefficiencies in read and write operations due to mismatched consistency levels between write and read operations, leading to potential query failures and excessive resource consumption, especially when replicas are unavailable.
Innovation Solution
Implementing an any-local-quorum consistency level that allows operations to proceed with a successful response from one data center, paired with an each-quorum consistency level on the opposing operation, ensuring data is written to a quorum of replica nodes and found in one data center, thereby optimizing performance and resource use.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a quorum consistency level is used for read operations across all replicas, then data availability is improved, but operation time and resource consumption increase significantly
Solution Approach 1:
The patent segments the distributed database into multiple data centers, each with its own local quorum. Instead of requiring a quorum across all replicas globally, the system allows read operations to succeed if a quorum is achieved in any single data center. This segmentation enables faster reads by limiting the scope of quorum verification to local data centers rather than the entire distributed system.
Solution Approach 2:
The patent applies partial action by requiring quorum only in one data center rather than in all data centers. This partial quorum approach is sufficient for read operations because it guarantees data availability while reducing the number of nodes that need to respond, thereby decreasing operation time and resource consumption.
2Productivity
If a low consistency level is used for write operations, then write speed is improved, but data reliability decreases
Solution Approach 1:
The patent segments the write operation requirement across data centers. Instead of requiring a fixed number of replicas globally, the system requires that a quorum be achieved in each data center where data is written. This segmentation allows writes to proceed quickly in each local data center while maintaining reliability through the quorum requirement at each segment.
Solution Approach 2:
The patent applies different consistency requirements to different locations (data centers). Each data center enforces a local quorum for writes, ensuring that data is reliably stored within that data center. This local quality approach allows the system to maintain high reliability without requiring global consistency, thereby improving write speed.
3Reliability
If all replicas are queried for read operations, then data availability is guaranteed, but resource consumption increases excessively
Solution Approach 1:
The patent extracts the quorum requirement from the global level and applies it only at the data center level. Instead of querying all replicas across all data centers, the system extracts and applies the quorum condition independently in each data center. This extraction reduces the number of queries needed while maintaining data availability guarantees.
Solution Approach 2:
The patent uses partial action by querying only enough data centers to achieve one successful quorum rather than querying all data centers. This partial approach is sufficient to guarantee data availability while significantly reducing the number of nodes queried, thereby lowering resource consumption.
4Reliability
If inter-data center communication is increased to ensure consistency, then data reliability is improved, but operation speed decreases
Solution Approach 1:
The patent segments the consistency requirement so that each data center maintains its own local quorum independently. This segmentation reduces the need for extensive inter-data center communication because each data center can verify consistency locally within its own quorum, rather than requiring coordination across all data centers.
Solution Approach 2:
The patent applies preliminary action by establishing local quorums in each data center before cross-data center operations. This preliminary local verification ensures that data is available and consistent within each data center, reducing the need for subsequent communication and coordination between data centers, thereby improving operation speed.
Data Source
AI summary
A storage application is disclosed herein that employs a method for managing read and write requests associated with a database distributed across multiple replicas in multiple data centers. The storage application may be implemented on one or more computing devices in the context of a data center, cloud environment, or the like. The method includes the storage application receiving a client request to perform an operation with respect to an object at an any-local-quorum consistency level. In response to client request, the storage application iteratively generates and submits coordinator requests to the multiple data centers at a local-quorum consistency level, until one of the multiple data centers successfully replies to one of the coordinator requests with a result. The storage application then replies to the client request based on the result provided by the one of the multiple data centers.