Backup Data Block Prioritization via Criticality Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional block-based backup systems fail to prioritize data blocks effectively, leading to potential data loss during disasters as critical data is queued behind non-critical data due to a first-come-first-served streaming approach, which increases the recovery point objective (RPO) and compromises service-level agreements (SLAs).
Innovation Solution
A method that assigns a weighted value to data blocks based on their criticality, calculated using Euclidean distances, to prioritize the streaming of data blocks during backups, ensuring that critical data is written to the target device first by appending criticality information to data block headers and sorting blocks accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all data blocks are streamed to the target device in first-come-first-served order, then the backup process is simple and fast, but critical data may be lost during disasters as non-critical data is written first
Solution Approach 1:
The patent applies local quality by assigning different priority levels to different data blocks based on their criticality. Each data block is tagged with a priority indicator (P0, P1, P2, P3) that reflects its importance, allowing the system to treat critical data differently from non-critical data during the backup process. This resolves the contradiction by maintaining simple first-come-first-served streaming for non-critical data while ensuring critical data receives priority treatment through the priority indicator mechanism.
Solution Approach 2:
The patent changes the parameter of data block priority by introducing a priority indicator parameter that can take different values (P0-P3). This parameter is appended to each data block header, transforming the uniform treatment of all data blocks into a differentiated treatment based on criticality. The priority indicator enables the system to dynamically adjust which data blocks are written first without fundamentally changing the backup architecture, thus improving reliability while maintaining relatively simple system complexity.
2Reliability
If critical data is prioritized for streaming, then data loss risk is reduced, but the backup system becomes more complex with priority assignment and sorting mechanisms
Solution Approach 1:
The patent applies preliminary action by determining and tagging the priority indicator for each data block before the actual backup streaming begins. The source host agent analyzes data objects, determines their criticality, and appends the appropriate priority indicator (P0-P3) to each data block header in advance. This preliminary prioritization allows the target device to simply follow the priority indicators during streaming without complex real-time decision-making, thus improving recovery point objective while keeping the prioritization mechanism relatively simple.
Solution Approach 2:
The patent introduces a priority indicator as an intermediary element that mediates between the source host and target device. Instead of the target device needing to complexly evaluate and prioritize data blocks in real-time, the priority indicator serves as a pre-computed intermediary that carries the prioritization information throughout the backup process. This intermediary simplifies the target device's operation while ensuring critical data receives appropriate priority treatment, resolving the contradiction between reliability and complexity.
3Productivity
If multiple heterogeneous data blocks are queued simultaneously, then backup throughput is maintained, but critical data waits behind non-critical data reducing backup priority effectiveness
Solution Approach 1:
The patent applies local quality by assigning different priority indicators to different data blocks within the same backup stream. Even when multiple heterogeneous data blocks are queued simultaneously to maintain high throughput, each block carries its own priority indicator (P0-P3) that reflects its criticality. This allows the system to process data blocks in parallel while ensuring that critical data blocks are identified and can be prioritized for writing when the target device is ready, thus maintaining both productivity and reliability.
Solution Approach 2:
The patent introduces dynamic prioritization where the priority indicator for each data block can change based on its criticality assessment. The source host agent dynamically determines the priority level for each data object and tags it accordingly before streaming. This dynamic tagging allows the system to adapt to different data types and criticalities in real-time, ensuring that critical data is always identified properly even when multiple data blocks are being processed simultaneously, thus maintaining both throughput and priority effectiveness.
Data Source
AI summary
Methods and systems for block-level data prioritization during a backup operation are disclosed. According to some embodiments, the method includes extracting a backup data criticality from header information of one or more data blocks. The method further includes based on the extracted backup data criticality, assigning a weighted value corresponding to the backup data criticality. The method further includes for each data block, calculating a Euclidean distance of the data block to a consecutive data block using the weighted value.