Attribute Model-Based Compliance Monitoring for Stored Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face inefficiencies in managing storage drives, particularly in flash-based systems, due to redundant write operations and lack of direct control by the operating system, leading to increased latency and reduced reliability.
Innovation Solution
Implementing a direct-mapped flash storage system where the operating system directly addresses data blocks without translation by storage controllers, offloading device management tasks to the storage array controllers, and utilizing non-volatile RAM for quick data buffering to enhance write latency and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If storage controllers translate data block addresses, then storage operations can be performed, but write latency increases and reliability decreases
Solution Approach 1:
The patent extracts the address translation function from the storage controller and implements it directly in the operating system kernel. This removes the translation overhead from the storage I/O path, allowing writes to reach the flash memory directly without controller intervention, thereby reducing latency and improving reliability.
Solution Approach 2:
The patent introduces a direct-mapped flash storage layer as an intermediary between the operating system and the flash memory devices. This layer provides direct block access without requiring controller translation, effectively mediating between the OS and storage hardware to eliminate translation latency.
2Productivity
If storage controllers manage device operations, then storage operations can be performed, but redundant write operations increase
Solution Approach 1:
The patent extracts device management responsibilities from the storage controller and transfers them to the operating system. The OS directly manages flash memory blocks, eliminating redundant write operations caused by controller-level management overhead and wear leveling.
Solution Approach 2:
The flash storage system implements self-service through direct OS control, where the operating system independently manages data block allocation and writes without requiring controller intervention. This self-managed approach eliminates redundant writes and reduces wear on flash memory components.
3Speed
If direct-mapped flash storage is implemented, then write latency improves, but device management complexity increases
Solution Approach 1:
The patent implements a universal address space where the operating system directly addresses flash memory blocks using standard block device interfaces. This multi-functional approach allows the same OS mechanisms used for other storage devices to work directly with flash memory, simplifying management despite the direct-mapped architecture.
Solution Approach 2:
The direct-mapped flash storage layer acts as an intermediary that presents a simplified interface to the operating system while handling the complexity of direct flash access. This mediator absorbs the management complexity, allowing the OS to interact with flash storage using standard, familiar mechanisms.
Data Source
AI summary
An illustrative method includes determining, by a monitoring system based on an attribute model of a dataset stored within a storage system, that one or more data items in the dataset are noncompliant with a compliance ruleset, wherein one or more data attributes of a data item in the dataset includes one or more of a data type of the data item, an encryption information of the data item, an accessibility of the data item, a retention policy of the data item, or an operation log configuration associated with the data item; determining, by the monitoring system based on the one or more data items being noncompliant with the compliance ruleset, a compliance level of the dataset with the compliance ruleset; and performing, by the monitoring system based on the compliance level of the dataset satisfying a compliance level threshold, an operation with respect to the dataset.