Главная
Study mode:
on
1
Intro
2
Background
3
Malware data mining is useful
4
Pre-BinaryPig: Architecture
5
BinaryPig - Results Exploration
6
BinaryPig Loaders
7
Optimizations in BinaryPig
8
BinaryPig: Loader Implementations
9
BinaryPig: Scripting
10
Web Interface
11
General Findings
12
Feature Extraction
13
Feature Depth
14
Clustering Results **.
15
ICO Extraction
16
Icon Features
17
Lessons Learned
18
Future work
19
black hat USA 2013
Description:
Explore scalable malware analytics using Hadoop in this Black Hat USA 2013 conference talk. Learn how Endgame developed BinaryPig, an open framework built on Apache Hadoop, Apache Pig, and Python, to process and analyze massive amounts of malware data. Discover techniques for handling terabytes of binary data, extracting feature sets for machine learning, and performing large-scale malware studies. Gain insights into the challenges of processing millions of malware samples and how BinaryPig addresses issues of scalability, workflow development, and parallel processing. Examine the architecture, optimizations, and implementations of BinaryPig, including loaders, scripting, and web interface. Delve into general findings, feature extraction methods, clustering results, and icon analysis. Understand the lessons learned and future directions for scalable malware analytics in the face of ever-increasing data volumes.

BinaryPig - Scalable Malware Analytics in Hadoop

Black Hat
Add to list