Cloud Data Protection Storage via Buffering and Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional data protection systems face challenges in supporting various cloud-based storage models, leading to increased development costs and slow deployment due to the need for custom interfaces for each storage service, which may not allow random access writes and require full sequential access to data.
Innovation Solution
The system submits buffered portions of a data protection image to a third-party storage service as single data objects and maintains data maps to associate the streamed data with storage objects, allowing for flexible buffering and write-once updates, enabling access to diverse storage models without extensive configuration changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If custom storage interfaces are used for each cloud-based storage service, then data protection functionality can be achieved, but development costs increase and deployment slows down
Solution Approach 1:
The patent implements a universal storage interface that can interact with multiple cloud-based storage services (object storage, file storage, block storage) through a single standardized API. This interface abstracts the underlying storage model differences, allowing the same data protection system to work across different cloud providers without requiring custom interfaces for each service type.
Solution Approach 2:
The patent introduces a cloud storage service adapter as an intermediary layer between the data protection system and various cloud storage services. This adapter translates data protection operations into cloud-specific APIs, enabling the system to communicate with different storage models without exposing complexity to the user.
2Quantity of substance
If cloud-based storage services are used for data protection, then storage capacity and accessibility improve, but support for varied storage models becomes more complex
Solution Approach 1:
The patent dynamically adjusts buffer sizes and data chunking parameters based on the characteristics of the target cloud storage service. The system can change buffering parameters such as buffer size, chunk size, and data segmentation strategy to optimize performance for different storage models (object storage vs. file storage vs. block storage).
Solution Approach 2:
The patent implements dynamic adaptation where the data protection system can switch between different data protection methods (copy, incremental backup, differential backup) based on the capabilities and characteristics of the cloud storage service being used. This allows the system to optimize for each storage model while maintaining a unified interface.
3Productivity
If sequential access is required for cloud storage, then data protection can be implemented, but random access capabilities are limited
Solution Approach 1:
The patent segments data into smaller chunks or blocks that can be processed and uploaded in parallel to the cloud storage service. This segmentation allows the system to maintain sequential processing for data protection while enabling random access to individual segments or blocks in the cloud storage, improving operational flexibility.
Solution Approach 2:
The patent introduces a virtualization layer that maps sequential data protection operations to random access operations in the cloud storage dimension. Through this dimensional translation, the system can perform sequential data protection processing while accessing data randomly in the cloud storage space, resolving the access pattern conflict.
4Productivity
If buffer sizes are increased to improve data protection efficiency, then processing speed increases, but memory requirements and data loss risk increase
Solution Approach 1:
The patent implements variable buffer sizes that can be dynamically adjusted based on the data protection method, cloud storage service characteristics, and system resources. Instead of using a fixed large buffer, the system uses partial buffering appropriate to each scenario, optimizing between processing speed and memory consumption.
Solution Approach 2:
The patent maintains continuous data protection processing through efficient buffer management that ensures data flows continuously from the source to the cloud storage service. The buffering mechanism is designed to minimize interruptions while maintaining optimal buffer sizes, ensuring continuous useful action without excessive memory consumption.
Data Source
AI summary
A computer-implemented method for cloud-based data protection storage may include (1) identifying a request to maintain a data protection image for a volume of data, (2) identifying a third-party storage service to use for storing the data protection image, (3) receiving a stream of data to apply to the data protection image, (4) buffering a portion of the stream of data, (5) writing the buffered portion of the stream to the third-party storage service as a single storage object, and (6) associating the single storage object with the buffered portion of the stream in a data map. Various other methods, systems, and computer-readable media are also disclosed.