Level Two Cache Raw Data Storage for Security and Integrity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional caching systems face challenges in managing raw and cooked data, particularly in persistent storage mediums like SSDs, where storing cooked data violates security standards and checksum mechanisms are not applicable, and encryption/decryption at the cache level is not feasible due to missing security keys.
Innovation Solution
The system stores both raw and cooked versions of data blocks in a level-one cache, with only the raw version being copied to a persistent level-two cache, maintaining security integrity and extending checksum computations to the level-two cache, allowing for efficient data management and access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If cooked data is stored in persistent storage medium (level-two cache), then data is readily accessible, but security integrity is violated
Solution Approach 1:
The patent divides data into two distinct versions: raw version for persistent storage and cooked version for fast access. This segmentation allows each version to serve its specific purpose - the raw version maintains security integrity for compliance, while the cooked version provides rapid access for performance-critical operations.
Solution Approach 2:
The level-one cache acts as an intermediary between the raw data in persistent storage and the cooked data needed for fast access. It stores both raw and cooked versions, enabling the system to read cooked data quickly while writing only raw data to persistent storage, thus resolving the contradiction between speed and security.
2Reliability
If encryption/decryption is performed at cache level, then data security is maintained, but system complexity increases due to missing security keys
Solution Approach 1:
The patent extracts the encryption/decryption operation from the cache level and relocates it to the storage level. Security keys are stored with the raw data in persistent storage rather than in the cache, eliminating the complexity of managing security keys at the cache level while maintaining data security integrity.
3Reliability
If checksum mechanisms are applied to cooked data, then data integrity is verified, but the mechanism becomes inapplicable for persistent storage
Solution Approach 1:
The patent segments data into raw and cooked versions with distinct integrity verification mechanisms. Checksums are computed and stored with raw data in persistent storage, while cooked data in the cache has its integrity verified through different means, making each mechanism applicable to its specific data type.
Solution Approach 2:
The level-one cache serves as an intermediary that handles the transition between raw data with checksums and cooked data with alternative verification. It maintains both versions and manages the integrity verification processes appropriately for each, ensuring checksum mechanisms remain applicable where needed.
4Productivity
If cache capacity is increased to store more data, then data reuse probability improves, but cost increases due to limited cache size
Solution Approach 1:
The patent introduces a new dimension to cache management by storing both raw and cooked versions of data. This effectively doubles the usable data capacity for a given physical cache size, as the cache can utilize both versions of data blocks, thereby improving data reuse efficiency without proportionally increasing hardware cost.
Data Source
AI summary
Aspects of the present disclosure disclose systems and methods for managing a level-two persistent cache. In various aspects, a solid-state drive is employed as a level-two cache to expand the capacity of existing caches. Any data stored in the level-two cache may be stored in a particular version or format of data known as “raw” data, in contrast to storing the data in a “cooked” version, as is typically stored in a level-one cache.