Tape Block to Object Mapping for Storage Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Object-based storage systems face delays in data retrieval due to the need to access remote locations, while tape-based storage systems suffer from sequential access latencies, and there is a lack of efficient integration of tape drive capabilities with object-based systems for handling long delays.
Innovation Solution
A data management engine maps tape blocks to objects in an object-based storage system, allowing for the conversion of tape drive commands into object-based storage operations, enabling efficient storage and retrieval of data streams in a tape archive format, thereby bridging the gap between tape-based and object-based storage systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is stored in an object-based storage system, then data can be accessed directly without sequential access delays, but retrieval delays occur due to the need to access remote locations
Solution Approach 1:
The system performs preliminary actions by pre-fetching data from remote locations before it is actually requested. When data is stored in the object-based storage system, the system proactively retrieves data that is likely to be needed soon, storing it in local cache memory. This preliminary action eliminates retrieval delays when the data is actually requested, as it is already available locally.
Solution Approach 2:
The patent introduces an intermediary component - a cache memory system - that sits between the object-based storage system and the user/application. This intermediary pre-loads data from the remote object-based storage into local cache, mediating the access delay by providing fast local access to frequently needed data while maintaining the direct access capability of object-based storage.
2Reliability
If tape-based storage is used, then data stability is achieved for archival purposes, but sequential access latencies prevent efficient data retrieval
Solution Approach 1:
The system segments data access into two distinct paths: frequently accessed data is retrieved quickly from local cache memory, while less frequently accessed data remains in the stable but slower object-based storage system. This segmentation allows the system to maintain data stability in the primary storage while providing fast access for active data through the cache layer.
Solution Approach 2:
The system performs preliminary data retrieval from the stable object-based storage system and caches it locally before actual access is needed. This preliminary action transfers data from the stable but slow storage medium to fast local memory in advance, eliminating sequential access latencies when the data is actually requested while maintaining the reliability of the primary storage system.
3Reliability
If multi-tiered storage systems are implemented, then data can be stored with high reliability, but complex retrieval processes create delays
Solution Approach 1:
The patent extracts the complexity of multi-tiered storage management by isolating it into a dedicated cache management system. Rather than implementing complex multi-tiered storage throughout the entire system, the solution extracts and implements a simpler two-layer system: object-based storage for reliability and a local cache layer for speed. This extraction reduces overall system complexity while maintaining the benefits of tiered storage.
Solution Approach 2:
The system performs preliminary data movement from the reliable object-based storage to local cache memory based on predicted access patterns. This preliminary action simplifies the retrieval process by pre-positioning data in the fastest storage layer available, eliminating the need for complex real-time decisions about where to retrieve data from during actual access operations.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Operations include storing and/or accessing data in tape volume containers. A system receives a request to access a tape block in a data storage tape. The system maps the requested tape block in the data storage tape to an object in a container. The system then retrieves data from the object in the container. The system generates, from the retrieved data, a data stream in tape archive format. The system transmits the data stream, in tape archive format, in response to the request to access the tape block in the data storage tape.