defenseunicorns / uds-core

A FOSS secure runtime platform for mission-critical capabilities
https://uds.defenseunicorns.com
GNU Affero General Public License v3.0
50 stars 21 forks source link

Evaluate Tetragon as a NeuVector replacement #888

Open mjnagel opened 3 weeks ago

mjnagel commented 3 weeks ago

https://tetragon.io/

Evaluate:

Definition of done: A scorecard evaluating the above criteria.

rjferguson21 commented 3 hours ago
Criteria Tetragon NeuVector
Pros - Minimalistic - Full end-to-end security, including runtime, network, and vulnerability protection
- Supported by Cisco/Isovalent - Network segmentation with zero-trust enforcement
- Performance - Enterprise-ready with compliance, vulnerability scanning, and firewall features
- Kubernetes Native - Integrated enforcement, blocking malicious behavior in real-time
Cons - No visual UI - Heavier resource consumption due to the extensive feature set
- Would require education on crafting policies - Complexity may overwhelm smaller deployments
- Missing L7 network security - Full enterprise features require a commercial version
Community Adoption Popular on GitHub with 3600 stars but unclear about enterprise adoption Enterprise focus, strong post-acquisition support by SUSE; less adoption in the general open-source community
Chainguard / IronBank Support IronBank images exist, Chainguard images not published NeuVector is supported in IronBank (DoD-approved hardened container registry), suitable for government use
Runtime Detection Provides runtime detection via eBPF Behavioral learning with runtime detection for container processes, files, and network traffic
Network Security Provides network security via eBPF, significantly lower level than Neuvector (using kprobes / tracepoints) Provides network-layer security, including segmentation, DPI (Deep Packet Inspection), and zero-trust policies
Compliance & Vulnerability Scanning N/A Built-in vulnerability scanning, compliance checks for PCI-DSS, GDPR, NIST, HIPAA
Enforcement Tetragon allows enforcing events in the kernel inline, either via SIGKILL or overriding return value of function Native enforcement engine to block network, process, and container-level threats in real-time
Out-of-the-Box Experience Much more primitive tool compared to Neuvector, easy to setup. Not a lot of policy examples. Comprehensive out-of-the-box security, with network policies, vulnerability scans, and compliance reports pre-configured
Helm / Config as Code Alignment Official Helm chart available, supports configuration via Kubernetes CRDs Helm chart available, configuration-as-code support with advanced controls; more complex due to the breadth of features
Ease of Use / Learning Curve Very steep learning curve, no UI Steeper learning curve due to the full feature set (network, compliance, runtime) but offers behavioral learning to automate setup
Resource Consumption Seems very performant and resource conscious Resource-intensive, especially with network security, DPI, and multi-layer enforcement
Integration with CI/CD N/A Integrates well with CI/CD tools, can block non-compliant containers at runtime and during image scans
Policy Management Requires authoring custom policies using Tetragon CRDs Automated policy learning with ability to create manual policies, suitable for both network and runtime security
Enterprise Support / Features No specific enterprise version for Tetragon, might have enhanced functionality with Cillium Full enterprise support, including multi-cloud deployments, built-in compliance management, and security dashboards
Service Mesh Compatibility Service Mesh compatible but no direct integration Provides strong integration with service meshes, enforcing network security and segmentation policies within service mesh traffic
Logging and Reporting Supports prometheus metrics and logging of Tetragon events. Built-in real-time dashboards for network and runtime events; advanced reporting for compliance and threat detection
Cloud Provider Compatibility Works across all major cloud providers (AWS, GCP, Azure) in any Kubernetes environment (self-managed or managed) Optimized for multi-cloud (AWS, GCP, Azure) and hybrid environments; certified for key public cloud providers and supports multi-cluster setups