Decoupling Cluster Data via External Database Persistence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In cloud computing environments, the tight coupling between clusters and their data prevents users from suspending clusters without losing data or incurring unnecessary charges, leading to potential data loss during unexpected shutdowns.
Innovation Solution
Decoupling cluster data from cloud deployment by persisting it in an external database, allowing clusters to be taken down and re-deployed with preserved data, and aligning persisted topologies with current cluster configurations for resumed execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cluster data is tightly coupled with cloud deployment, then data is preserved during cluster operations, but cluster suspension is prevented and unnecessary charges continue to accrue
Solution Approach 1:
The patent segments the coupling between cluster and data by introducing an external database as an intermediary storage layer. Cluster data is separated from the cloud deployment environment and stored independently in the external database, allowing the cluster to be suspended or terminated without affecting data persistence. This segmentation resolves the contradiction by enabling independent management of compute resources and data storage.
Solution Approach 2:
The external database acts as an intermediary between the cluster and the data storage function. Instead of direct coupling where data resides within the cluster environment, the external database mediates data access and persistence. This intermediary layer allows the cluster to be taken down while data remains preserved in the external database, eliminating the contradiction between data safety and suspension capability.
2Reliability
If cluster is kept online to preserve data, then data is accessible, but resource costs continue to accrue unnecessarily
Solution Approach 1:
The patent extracts the data storage function from the running cluster environment and places it in an external database. This extraction allows data to be preserved and accessible independently of whether the cluster is running or suspended. Users can terminate the cluster to stop incurring compute resource charges while data remains safely stored and accessible in the external database when needed.
Solution Approach 2:
The system changes the state parameter of data storage from 'coupled with running cluster' to 'independent external storage'. This parameter change enables data to maintain accessibility (a state property) while the cluster transitions between running and suspended states, thereby eliminating continuous resource costs while preserving data availability.
3Productivity
If unexpected shutdown occurs with tight coupling, then cluster operation stops, but data is lost
Solution Approach 1:
The patent implements beforehand cushioning by pre-configuring an external database as a safety buffer for cluster data. Before any cluster shutdown (expected or unexpected) can occur, data is already secured in the external database environment. This prior cushioning ensures that even if the cluster experiences unexpected failure, data loss is prevented because data never resided solely in the volatile cluster environment but was continuously persisted externally.
Data Source
AI summary
The present invention extends to methods, systems, and computer program products for decoupling cluster data from cloud deployment. The decoupling facilitates taking down and reattaching a cluster to persisted data as needed. Other embodiments of the invention include resuming a cluster's execution on a different topology and/or different system states based on a persisted data set.