Applying artificial intelligence to database security
DB Networks patented technology is based on deep protocol extraction, machine learning, and behavioral analysis. As a result you'll gain real-time and continuous situational awareness of your database infrastructure.

Simplified, transparent, drop-in installation

Non-intrusively discovers databases

Offers detailed insights into the interactions between applications and their connected databases

Detects database infrastructure policy violations

Pinpoints compromised credentials in real-time

Rapid behavioral model construction

Immediate database protection from potentially vulnerable legacy applications and 3rd party applications

Identifies database attacks (including Zero-Day attacks) through artificial intelligence - no signatures required

Immediate database protection against application framework vulnerabilities

Implemented as physical or virtual appliance

Home / Technology
DB Networks Technology
Field Proven and Award Winning Database Security Technology

While Web applications can produce dynamic, and often extremely complex SQL, it turns out the application behavior when SQL is created can be modeled. DB Networks technology applies deep protocol extraction to all SQL statements that have been created by the database application clients. DB Networks then uses artificial intelligence based on machine learning to construct a unique multi-dimensional behavioral model of each application. Using this unique behavioral model, each SQL statement is then subjected to a thorough lexical analysis and SQL semantic analysis. Any rogue SQL statements are immediately identified and the organization's pre-defined alarm procedure is invoked. This approach has proven to be extremely accurate in many of the world's largest database infrastructures.

Signature-less Database Security

Our technology is a novel and patented approach that actually learns and models an applications unique behavior for generating its SQL statements. After the short machine-learning period, required to construct the unique behavioral model, a suite of detection algorithms evaluates each SQL statement against the application's unique behavioral model. New SQL statements, not seen during the machine learning process, go through comprehensive structural analysis. Any SQL statements not consistent with the established behavioral model are identified as likely attacks. This entire process is entirely automatic. DB Networks technology is completely plug-and-play -- requiring no blacklists, signature files, or whitelists to be configured or maintain. Artificial intelligence has proven to be rapid and extremely accurate at identifying even then stealthiest of database attacks.

Technical Requirements

  • Oracle server release 8i (8.1.7) or later
    Microsoft SQL Server version 2000 or later
  • Bi-directional mirrored port or passive
    tap capture to feed 10/100/1000 Mbit/sec capture ports
  • IDS-6300v virtual appliance supported
    under VMware ESXi 5.1, 5.5

System Specifications

  • 2U x 19 inch rack mount form factor
  • Dual redundant power supplies -300W
  • 2 TB of RAID10 storage for captured workloads
  • 480 GB High performance SSD
  • 2 TB archival storage
  • Encrypted data
  • Operator authentication
  • Role based permissions to limit access to sensitive data
  • Support for encrypted database interfaces