Cloud Storage Access Layer for Malicious Pattern Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud infrastructure faces significant challenges in protecting non-volatile storage from malicious attacks, which can compromise data integrity and accessibility, affecting multiple applications and services shared on the same storage infrastructure.
Innovation Solution
The implementation of a Dynamic Adaptive Learning (DAL) layer that maintains and updates lists of malicious patterns, scans access requests for these patterns, and uses AI/ML to identify new attacks by analyzing error codes, thereby blocking malicious requests and preventing disruptions to other applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If non-volatile storage is shared among multiple applications in cloud infrastructure, then resource utilization and scalability are improved, but vulnerability to malicious attacks and data integrity risks increase
Solution Approach 1:
The patent introduces a storage access layer as an intermediary component between applications and non-volatile storage. This layer implements security functions including malicious pattern detection, access request filtering, and data integrity verification. By placing this intermediary layer, the system enables shared storage functionality while protecting against malicious attacks that could compromise data integrity across multiple applications.
2Reliability
If security checks are implemented on storage access requests, then data integrity protection is improved, but access request processing time increases
Solution Approach 1:
The patent implements preliminary action by pre-computing and maintaining malicious patterns in a cache structure before actual storage operations occur. The system pre-processes security rules and patterns, so that during runtime, access requests can be checked against these pre-prepared patterns using efficient string matching algorithms. This reduces the processing time during actual storage operations while maintaining comprehensive security checks.
Solution Approach 2:
The patent applies local quality by implementing selective security checking based on the specific characteristics of access requests. The storage access layer identifies and applies relevant malicious patterns specific to each request type and storage operation. Rather than uniformly checking all requests with all patterns, the system locally optimizes by matching requests with appropriate security checks, reducing overall processing time while maintaining data integrity protection.
3Measurement precision
If comprehensive malicious pattern detection is implemented, then attack identification accuracy is improved, but computational complexity and resource consumption increase
Solution Approach 1:
The patent segments the security detection system into distinct functional components: pattern generation module, pattern caching layer, and pattern matching engine. Each component handles specific aspects of malicious pattern detection. The segmentation allows the system to maintain comprehensive detection capabilities while managing complexity through modular design, where each segment can be independently optimized and maintained.
Solution Approach 2:
The patent uses copying by maintaining cached copies of malicious patterns in memory rather than repeatedly accessing and processing the original comprehensive pattern databases. The system creates and stores replicated pattern structures that can be quickly referenced during access request validation. This copying approach reduces computational complexity during detection operations while preserving the accuracy of comprehensive pattern matching.
Data Source
AI summary
Protecting data in non-volatile storages provided to clouds against malicious attacks. According to an aspect, multiple malicious patterns indicating respective malicious attacks to access non-volatile storages provided to clouds in a cloud infrastructure are maintained. When an access request is received, the data stream representing the access request is examined to determine whether the data stream contains any of the malicious patterns. If the data stream is found not to contain any malicious pattern, it is concluded that the access request is free of the malicious attacks. If the data stream is found to contain at least one malicious pattern, it is concluded that the access request is a malicious attack corresponding to the malicious pattern.