Content Node Encrypted Purging via Segmented Key Erasure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Content delivery networks face challenges in efficiently managing and purging cached content from content nodes, leading to server overload and slow performance due to frequent user requests, as existing methods do not effectively handle updates or removals of cached data without overwriting actual content objects.
Innovation Solution
Implementing a method to encrypt data objects using separate encryption keys with a common and individualized portion, allowing for selective purging of these keys to erase content without decrypting remaining objects, thereby preventing access and reducing the need for writes to storage devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If content is cached in content nodes to reduce server load, then user access speed is improved, but content updates and purges become complex and inefficient
Solution Approach 1:
The encryption key is segmented into two distinct portions: a common portion shared by multiple data objects and an individualized portion unique to each data object. This segmentation enables selective purging operations where only specific key portions need to be modified to achieve content removal, simplifying content management while maintaining caching benefits.
Solution Approach 2:
The patent extracts the purging functionality from complex content manipulation operations by implementing key portion erasure. Instead of overwriting or moving content objects, the system simply erases the relevant encryption key portions, making the purging operation straightforward and efficient while content remains in cache for ongoing access.
2Loss of information
If traditional purging methods overwrite content objects, then content is removed from cache, but storage writes increase and performance decreases
Solution Approach 1:
The patent removes the harmful overwriting operation entirely by extracting only the essential purging function: erasing encryption key portions. This approach achieves complete content removal from cache while avoiding unnecessary storage writes to content objects, thereby maintaining storage performance.
Solution Approach 2:
Instead of applying a uniform purging operation that overwrites entire content objects, the patent applies a localized operation that only erases specific encryption key portions. This localized approach achieves the desired content removal effect without the performance penalty of writing to storage devices.
3Reliability
If all content is purged from content nodes, then outdated content is removed, but frequent purging operations increase server load
Solution Approach 1:
The patent extracts the essential purging function to a simple key erasure operation that can be performed with minimal processing energy. By removing only the encryption key portions rather than manipulating content objects, the system maintains content freshness while significantly reducing the energy cost of purging operations.
Solution Approach 2:
The patent creates a logical copy of the content removal process by erasing key portions instead of physically removing content objects. This virtual purging mechanism maintains the effect of content removal without the computational overhead of actual content manipulation, reducing server energy consumption.
Data Source
AI summary
Described herein are methods, systems, and software for encrypting and erasing data objects in a content node. In one example, a method of operating a content node that caches content divided into one or more data objects includes encrypting the one or more data objects using separate encryption keys for each of the one or more data objects, the separate encryption keys comprising a common portion shared by the one or more data objects and an individualized portion unique to each data object. The method further provides receiving a purge request to erase at least one data object and, responsive to the purge request, erasing at least one of the common portion or the individualized portion for the at least one data object based on the purge request.