Partition-Scoped Snapshot Creation in Distributed Data Grids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing snapshot systems in distributed data grids require quiescing cache services, leading to prolonged unavailability and reduced service availability during snapshot processes, as all partitions must be made consistent at the same point in time.
Innovation Solution
Implementing partition-scoped snapshot creation, where snapshots are created one partition at a time, allowing cache service threads to remain active and only blocking write requests during snapshot creation for the specific partition, while read requests can continue and other partitions remain available.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional snapshot systems are used to ensure data consistency across all partitions, then snapshot consistency is improved, but cache service availability deteriorates due to quiescing requirements
Solution Approach 1:
The patent divides the snapshot operation into partition-level segments, allowing individual partitions to be snapshotted independently rather than requiring a global quiesce of all partitions. This segmentation enables the cache service to continue operating on non-snapshotting partitions, maintaining service availability while achieving snapshot consistency for each partition through controlled blocking only during its specific snapshot window.
2Stability of the object's composition
If all partitions are snapshotted simultaneously to ensure consistency, then snapshot atomicity is improved, but service unavailability duration increases
Solution Approach 1:
The patent implements periodic snapshot operations at the partition level, where each partition undergoes snapshot creation in periodic intervals rather than all at once. This approach maintains snapshot atomicity for each partition through controlled blocking while distributing the unavailability impact across different time periods, thereby reducing the total duration of service unavailability compared to simultaneous global snapshotting.
3Manufacturing precision
If cache service is quiesced during snapshot creation to ensure data consistency, then data consistency is improved, but partition availability deteriorates
Solution Approach 1:
The patent applies local quality by implementing partition-specific blocking only where and when needed for snapshot creation, rather than applying a global quiesce to all partitions. This allows the cache service to maintain full availability for reading and writing on partitions that are not currently being snapshotted, while ensuring data consistency for the specific partition undergoing snapshot through localized controlled blocking.
Data Source
AI summary
A system and method for partitioned snapshot creation of caches in a distributed data grid is provided. The system and method enables a snapshot to be created in a running system without quiescing a cache service. Moreover for each particular partition, execution of read/write requests are not blocked during the period that a snapshot creation task is being performed for the particular partition. The cache service thread continues to execute read requests for all partitions with write requests for the partition under snapshot experiencing delayed response. The system and method reduces the period of time for which partitions are unavailable during a snapshot process and increases the availability of cache services provided by a distributed data grid compared to prior snapshot systems.