Virtual Machine Antivirus Queue Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In virtual computing environments, simultaneous antivirus checking across multiple virtual machines can lead to performance issues due to overloading of the security virtual machine and redundant checks on identical files, especially during initial or update-related massive checks, where common databases are ineffective.

Innovation Solution

A system and method that involve forming a queue of identifiers for malware analysis on each virtual machine, determining a selection method based on the number of virtual machines performing analysis, and utilizing a security virtual machine to check for duplicate analyses, thereby optimizing the use of a common database and reducing redundant checks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by moving object

If a common database is used to store antivirus check results, then resource usage is optimized, but during initial or update-related massive checks, the database is ineffective because identical files from multiple virtual machines are checked redundantly

Engineering Contradiction:
Improveresource usageVSAvoidantivirus check efficiency
Core Design Contradiction:
Use of energy by moving objectVSProductivity

Solution Approach 1:

Virtual machines perform preliminary actions by sending file identifiers and hashes to the security virtual machine before actual antivirus scanning. The security virtual machine pre-processes these identifiers, checks the common database, and determines which files require scanning. This preliminary filtering action prevents redundant scans of identical files while maintaining resource optimization through the common database approach.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If multiple virtual machines simultaneously send files for antivirus checking, then checking coverage is improved, but the security virtual machine becomes overloaded and performance decreases

Engineering Contradiction:
Improvechecking coverageVSAvoidprocessing speed
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent extracts the essential identifying features of files (identifiers and hash values) from the actual file content and sends only these extracted elements to the security virtual machine for preliminary processing. This extraction reduces the data volume significantly, allowing the security virtual machine to handle multiple virtual machines simultaneously without overload, while still maintaining comprehensive checking coverage through subsequent scanning of identified files.

Inventive Principle:
Principle #2Taking out (Extraction)

3Reliability

If identical files from multiple virtual machines are checked simultaneously, then comprehensive coverage is achieved, but redundant checks increase overall processing time

Engineering Contradiction:
Improvechecking coverageVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system implements feedback mechanisms where virtual machines send file identifiers and hash values to the security virtual machine, which then checks the common database for existing results. Based on this feedback, the security virtual machine determines whether files require scanning or can be skipped. This feedback loop prevents redundant checks of identical files across multiple virtual machines, reducing processing time while maintaining comprehensive coverage through the centralized database.

Inventive Principle:
Principle #23Feedback

4Measurement precision

If all files from all virtual machines are scanned, then detection accuracy is improved, but the queue of requests overfills and virtual machines stand idle

Engineering Contradiction:
Improvedetection accuracyVSAvoididle time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

Virtual machines perform preliminary actions by submitting file identifiers and hash values before actual scanning. The security virtual machine uses these preliminary submissions to check the common database and identify files that require scanning versus those that have been previously checked. This preliminary filtering maintains detection accuracy for files that need scanning while reducing idle time by preventing unnecessary queue buildup from redundant scan requests.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3059692B1System and method for antivirus checking of objects from a plurality of virtual machines
Publication Date: 2017.04.05 AO KASPERSKY LAB
  • EP3059692B1 patent drawingFigure 1
  • EP3059692B1 patent drawingFigure 2
  • EP3059692B1 patent drawingFigure 3

AI summary

Disclosed are a system and a method for malware detection on virtual machines. An example method comprises: forming, on a virtual machine, a queue of identifiers of objects for malware analysis; determining a method for selecting objects in the queue for malware analysis; selecting one or more objects from the queue for malware analysis; providing identifiers of the selected objects to a security virtual machine for malware analysis; checking, by the security virtual machine, whether each of the selected objects has been previously provided for malware analysis by another virtual machine; when a selected object has not been previously provided by another virtual machine, performing, by the security virtual machine, a malware analysis of the selected object; and providing, to the virtual machine, a malware analysis result for the selected object.