Cartography: using graphs to improve and scale security decision-making
Talkat Cloud Native Security Day North America 2020 · on Tuesday, 2020-11-17· at 14:50· Virtual , VIrtual·
summary
A free open-source tool, Cartography pulls cloud assets into a neo4j graph and runs daily cron jobs in Kubernetes (statefulset for neo4j, cron job for Cartography) to model identities, roles and resources; a concrete finding shows developers able to assume a finance auditor role across accounts. 125 custom Cypher queries feed three dashboards in Kibana via an ingester to Elasticsearch and drive alerts to Slack for drift and misconfiguration; integration complements Terraform and proves practical at scale.
AI-written summary · about
chapters (18)
- 0:00Intro
- 1:03Moving too fast
- 1:44Modern infra is complicated!
- 3:14Limited resources
- 5:41We need a self-maintaining map
- 6:20Existing solutions
- 10:07Scenario: Okta to AWS transitivity
- 12:13Architecture
- 16:43Multi-Cloud Auditing
- 17:34Deployment on K8s
- 19:06Data Consumption - Neo4j
- 19:48Data Consumption - Query Format
- 20:47Data Consumption - Jupyter
- 21:38Elasticsearch Integration
- 22:55Data Consumption - Kibana
- 23:32High Level Dashboards
- 25:07Drift Detection
- 26:33Cartography's roadmap
related talks
28mFrom Side Project to Startup: Graph Lessons From the Cartography Journey
30mLyft Cartography: Automating Security Visibility and Democratization
29mGraphs: The Fabric of DevOps
24mYour company, as a Knowledge Graph - the foundation of cybersecurity’s future
20mAttack of the Graph: Visual Tools for Cyber Analysis
27mNeo4j's Risk Register: Maximise Your Data Insights
30mContext Engineering: Connecting the Dots with Graphs
23mLocation Insights for Your Graphs with Neo4j, PostGIS, and Brickgraph