WildFire reproduces a variety of analysis environments, including the operating system, to identify malicious behaviors within samples. Depending on the characteristics and features of the sample, multiple analysis environments may be used to determine the nature of the file. WildFire uses static analysis with machine learning to initially determine if known and variants of known samples are malicious. Based on the initial verdict of the submission, WildFire sends the unknown samples to analysis environment(s) to inspect the file in greater detail by extracting additional information and indicators from dynamic analysis. If the file has been obfuscated using custom or open source methods, the WildFire cloud decompresses and decrypts the file in-memory within the dynamic analysis environment before analyzing it using static analysis. During dynamic analysis, WildFire observes the file as it would behave when executed within client systems and looks for various signs of malicious activities, such as changes to browser security settings, injection of code into other processes, modification of files in operating system folders, or attempts by the sample to access malicious domains. Additionally, PCAPs generated during dynamic analysis in the WildFire cloud undergo deep inspection and are used to create network activity profiles. Network traffic profiles can detect known malware and previously unknown malware using a one-to-many profile match.
WildFire analyzes files using the following methods:
- Static Analysis—Detects known threats by analyzing the characteristics of samples prior to execution.
- Machine Learning—Identifies variants of known threats by comparing malware feature sets against a dynamically updated classification systems.
- Dynamic Unpacking (WildFire Cloud analysis only)—Identifies and unpacks files that have been encrypted using custom/open source methods and prepares it for static analysis.
- Dynamic Analysis—A custom built, evasion resistant virtual environment in which previously unknown submissions are detonated to determine real-world effects and behavior.
- Bare Metal Analysis (WildFire Cloud analysis only)—A fully hardware-based analysis environment specifically designed for advanced VM-aware threats. Samples that display the characteristics of an advanced VM-aware threat are steered towards the bare metal appliance by the heuristic engine.
WildFire operates analysis environments that replicate the following operating systems:
- Microsoft Windows XP 32-bit
- Microsoft Windows 7 64-bit
- Microsoft Windows 7 32-bit (Supported as an option for WildFire appliance only)
- Microsoft Windows 10 64-bit (WildFire Cloud Analysis only)
- Mac OSX (WildFire Cloud Analysis only)
- Android (WildFire Cloud Analysis only)
- Linux (WildFire Cloud Analysis only)
The WildFire public cloud also analyzes files using multiple versions of software to accurately identify malware that target specific versions of client applications. The WildFire private cloud does not support multi-version analysis, and does not analyze application-specific files across multiple versions.