Data Re-striping Across Striped Volume Sets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional re-striping techniques for data containers across multiple volumes are inefficient due to high computational overhead and require additional storage space, especially when volumes are added or removed, leading to performance degradation and storage inefficiencies.
Innovation Solution
A system and method for re-striping data containers across a striped volume set by creating new striping rules that incorporate additional volumes, allowing data relocation as a background process to maintain service efficiency and reduce storage requirements, with the option to use multiple sets of striping rules for new and existing data containers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional re-striping techniques are used to incorporate additional volumes, then storage capacity is increased, but computational overhead increases and performance degrades
Solution Approach 1:
The system performs re-striping operations as background processes before data containers are actually accessed, rather than during active use. This preliminary action allows the computationally intensive data relocation to occur when system resources are available, avoiding performance degradation during normal operations while still achieving the storage capacity expansion
Solution Approach 2:
The re-striping process is divided into separate background tasks that can execute independently and in parallel. Instead of performing a single large-scale re-striping operation that would overwhelm system resources, the data relocation is segmented into manageable chunks that can be processed incrementally without impacting overall system productivity
2Quantity of substance
If conventional re-striping techniques are used to incorporate additional volumes, then storage capacity is increased, but additional storage space is required for the process
Solution Approach 1:
The system creates incremental copies of data blocks only when necessary during the re-striping process, rather than requiring complete duplicate storage of all data. The copy-on-write mechanism allows the original data to remain on existing volumes while only the changed portions are copied to new volumes, significantly reducing the additional storage space required
Solution Approach 2:
The system dynamically adjusts the striping parameters and data block size based on the specific re-striping scenario. By changing parameters such as stripe width and block size, the system can optimize the amount of additional storage needed, allowing more efficient use of available space during the volume expansion process
3Speed
If re-striping is performed to maintain optimal data layout, then access efficiency is improved, but disruption to existing data containers occurs
Solution Approach 1:
The system performs re-striping operations in advance as background tasks, so that when data containers are accessed, the data is already in its optimal location. This eliminates the need to disrupt existing data containers during active access, maintaining both access efficiency and data stability simultaneously
Solution Approach 2:
The system applies re-striping selectively to only those portions of data containers that need to be relocated, rather than forcing a complete re-striping of entire containers. This local approach maintains the stability of unchanged data while optimizing the layout of relocated portions, preserving both reliability and access efficiency
Data Source
AI summary
A system and method re-stripes one or more data containers across a striped volume set (SVS) that has been modified by the addition of one or more volumes. The SVS is associated with an existing set of striping rules that define a stripe algorithm, a stripe width and an ordered list of volumes distributed across a plurality of nodes interconnected as a cluster. Each node of the cluster includes (i) a disk element (D-blade) adapted to service a volume of the SVS and (ii) a network element (N-blade) adapted to redirect a data access request to any D-blade of the cluster. Notably, the content of each data container is apportioned among the volumes of the SVS to thereby improve the efficiency of storage service provided by the cluster. To that end, the stripe algorithm specifies the manner in which the data container content is apportioned as stripes across the plurality of volumes, while the stripe width specifies the size/width of each stripe.