2.5.1. MetaDefender Core 4.19.0 or newer (PostgreSQL)

Updated: 27 Aug 2020

The current performance report is in BETA version.

OPSWAT continuously updates our performance report on regurlar basis.

Disclaimer: These results should be viewed as guidelines and not performance guarantees, since there are many variables that affect performance (file set, network configurations, hardware characteristics, etc.). If throughput is important to your implementation, OPSWAT recommends site-specific benchmarking before implementing a production solution.

System Specs

CPU

3.2GHz Core i7-8700

6 physical cores

12 logical cores

RAM

32 GB

HDD

SSD

150GB available disk

Deployment Setups

 

Deployment A

Deployment B

Deployment C

Deployment fashion

On-premise (Physical server)

On-premise (Physical server)

On-premise (Physical server)

MetaDefender Core version

4.18.0 (SQLite)

4.19.0 (PostgreSQL)

4.19.0 (PostgreSQL)

Database location

Local

Local

Remote

Platform

Windows

Windows

Windows

Licensing

8 AV engines

Deep CDR engine

8 AV engines

Deep CDR engine

8 AV engines

Deep CDR engine

  • Local: Database files managed in the same machine with MetaDefender Core

  • Remote: Database files managed in a different machine from MetaDefender Core, in local area network (LAN)

Data Set

  • Total data set size: 12.1 GB

  • Number of files: 2,000

    • File Category: Documents (MS Office, PDF)

    • Average file size: 6.2 MB

  • Total number of extracted files: 155,137

    • Average number of extracted files per one original file: 77.57

Test Scenarios

  • Number of simultaneous threads (for scan submission): 50

    • Always have 50 original document files active being processed on MetaDefender Core at any given time

  • Polling mode used for scan result retrieval

    • Interval time between every 2 polling requests: 200 ms

MetaDefender Core configurations:

  • Node queue size: 500

  • Archive handling settings:

    • Max allowed extracted files: 20000

    • Max allowed extracted size (GB): 2

    • Max allowed rescursive level: 5

    • Enabled to extract document files

  • Deep CDR settings: Enabled for all file types

  • AV scanning setting:

    • Engine timeout (minute): 1

    • Global scan timeout (minute): 10

Test Results

Run time: 45 minutes (2700 seconds)

 

Deployment A

(SQLite)

Deployment B

(PostgreSQL - Local)

Deployment C

(PostgreSQL - Remote)

Total successful extracted child files (objects) scanned

99,753

149,785

99,860

Total successful original document files scanned

1,266

1,928

1,304

Average processing speed (objects / sec)

36.94

55.48

36.99

Number of cancelled scan results (*)

57

0

5

Average number of polling requests needed for a complete successful scan on original document file (interval polling = 200 ms)

50.24

26.89

45.89

Notes:

(*): Cancelled result happens when after 5 minutes timeout passed without a complete final result, then client stopped polling and canceled the scan

images/download/attachments/4330543/image2020-8-27_13-42-49.png images/download/attachments/4330543/image2020-8-27_13-43-51.png

images/download/attachments/4330543/image2020-8-27_13-46-5.png images/download/attachments/4330543/image2020-8-27_13-47-11.png