Serverless Container Architecture for Secure Sensitive Data Stripping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional server-based systems for stripping sensitive data from files face security challenges and inefficiencies, particularly in scalability and resource management, leading to risks of unauthorized access and inefficiencies during varying workloads.

Innovation Solution

A serverless architecture is employed using cloud compute services that generate ephemeral container instances to process sensitive files, strip sensitive data, and transmit stripped files to storage, while terminating the container instances to ensure security and efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional server-based systems are used to strip sensitive data from files, then data processing can be performed continuously, but security risks increase due to persistent memory blocks and file pointers that may be accessed by unauthorized users

Engineering Contradiction:
ImprovesecurityVSAvoiddata processing capability
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent employs ephemeral container instances that are created temporarily to process sensitive files and then automatically terminated. These container instances exist only for the duration needed to strip sensitive data from a specific file, leaving no persistent memory blocks or file pointers that could be exploited by unauthorized users. This disposable approach eliminates the security vulnerability inherent in traditional persistent server-based systems while maintaining continuous data processing capability.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The system dynamically creates and terminates container instances based on the arrival of sensitive files requiring processing. Rather than maintaining static, persistent server resources, the system adapts its computing infrastructure in real-time, allocating resources only when needed and removing them immediately after use. This dynamic approach ensures that no sensitive data remains in memory or on disk after processing, resolving the contradiction between security and productivity.

Inventive Principle:
Principle #15Dynamics

2Productivity

If dedicated servers are allocated to process sensitive files continuously, then data stripping can be performed in near real-time, but resource inefficiency occurs during inactive periods with idle capacity

Engineering Contradiction:
Improvereal-time data processingVSAvoididle capacity costs
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

Instead of maintaining continuous, persistent server operations, the system implements periodic action by creating ephemeral container instances only when sensitive files arrive for processing. Each container instance is instantiated, processes the file, and is then terminated. This periodic creation and destruction of processing resources eliminates idle capacity during inactive periods while maintaining the ability to process files in near real-time when they arrive, resolving the contradiction between productivity and energy loss.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system automatically creates and terminates container instances based on the presence of sensitive files, without requiring manual resource allocation or management. When a sensitive file is detected, the system self-service provisions the necessary computing resources; when processing is complete, it automatically deallocates those resources. This self-service mechanism eliminates waste from idle capacity while ensuring real-time processing capability is maintained when needed.

Inventive Principle:
Principle #25Self-service

3Productivity

If traditional servers process large quantities of sensitive data, then high throughput is achieved, but security risk increases significantly if servers are compromised

Engineering Contradiction:
Improvevolume of data processedVSAvoidsecurity risk from server compromise
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent segments the data processing function into isolated, individual container instances, each handling a specific sensitive file. Rather than consolidating all data processing on a single persistent server, the system divides processing into numerous discrete, temporary container instances. This segmentation ensures that if one container is compromised, the attacker gains access only to that single file's data, not to a large volume of sensitive data stored on a persistent server, thus resolving the contradiction between processing volume and security risk.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

By using disposable, ephemeral container instances that are created and destroyed for each processing task, the system limits the window of opportunity for security breaches. Even if a container instance is compromised, the sensitive data exists in memory only temporarily during processing and is not persisted to disk or shared across multiple processes. This approach enables high throughput processing while minimizing the security risk associated with server compromise.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Data Source

PatentEP3842980B1System and method for automatically securing sensitive data in public cloud using a serverless architecture
Publication Date: 2023.05.31 CAPITAL ONE SERVICES LLC
  • EP3842980B1 patent drawingFigure 1
  • EP3842980B1 patent drawingFigure 2~4
  • EP3842980B1 patent drawingFigure 5A~5B

AI summary

Systems and methods are provided executing jobs immediately upon receipt of a notification. The systems and methods may include receiving, at a cloud compute service, a notification that a sensitive file comprising sensitive data has been received at a file receipt location, the sensitive file being sent by a client device; generating, by the cloud compute service, a container instance in response to the notification; retrieving, by the container instance, the sensitive file from the file receipt location; generating, by the container instance, a stripped file by stripping the sensitive data from the sensitive file based on a configuration file; transmitting, by the container instance, the stripped file to a storage location; deleting the sensitive file and associated file pointers from the file receipt location; and terminating the container instance, wherein terminating the container instance comprises deleting files comprising sensitive data and associated file pointers.