Filesystem Scanner for Storage Metadata Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage systems face challenges in efficiently distinguishing and caching filesystem metadata and data without modifying existing storage software or transport layers, particularly in scenarios where I/O hints are not embedded in requests, hindering the implementation of Differentiated Storage Services (DSS) and Cache Acceleration Software (CAS) techniques.
Innovation Solution
A filesystem scanner is implemented to detect and classify filesystem metadata on the host or remote storage side, generating classification tags for data, allowing the cache processor to distribute requests based on these tags, thereby enabling efficient caching without requiring modifications to the storage software stack.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If I/O hints are not embedded in requests, then compatibility with existing systems is maintained, but the ability to implement Differentiated Storage Services and Cache Acceleration Software is hindered
Solution Approach 1:
The filesystem scanner performs preliminary scanning and classification of filesystem metadata before I/O requests are processed. By pre-generating classification tags for metadata during a scanning phase, the system prepares caching information in advance, enabling efficient differentiated storage services without requiring modifications to the storage software stack or embedded hints in requests.
2Loss of time
If filesystem metadata is prioritized for caching, then I/O latency is reduced, but system complexity increases due to classification requirements
Solution Approach 1:
A filesystem scanner acts as an intermediary component between the host system and the cache processor. This scanner generates classification tags for filesystem metadata without requiring modifications to existing storage software or transport layers. The classification tags enable the cache processor to distinguish and prioritize metadata caching, reducing I/O latency while maintaining compatibility with existing systems through this intermediary layer.
Data Source
AI summary
A host computing arrangement is provided, which may include a host processor having a host operating system and host kernel associated therewith. The host processor may be configured to host a guest operating system, mirror a filesystem of the guest operating system via the host kernel, and generate caching criteria by scanning the mirrored filesystem. The host computing arrangement may further include a cache engine. The cache engine may be configured to process an I/O request from the guest operating system based on the caching criteria generated by the host processor.