# barrahome.org - LLM Context File ## Site Overview Personal blog and technical documentation by Alberto Ferrer. Focus: nginx, Linux system administration, performance optimization, distributed systems, and modern web architecture. ## Author - Name: Alberto Ferrer - Site: https://barrahome.org - GitHub: https://github.com/bet0x - LinkedIn: https://www.linkedin.com/in/bet0x/ ## Technology Stack - Hosting: nginx with custom ngx_markdown_filter_module - Content: Markdown files (.md) converted to HTML on-the-fly - No static site generator - real-time markdown rendering - Client-side tag filtering with vanilla JavaScript - Styling: Custom CSS with terminal/window aesthetic ## Content Structure ### Main Pages - Home: https://barrahome.org/ - Projects: https://barrahome.org/projects.md - CV/Resume: https://barrahome.org/cv.md - Contact: https://barrahome.org/contact.md ### Blog Posts (Latest First) - 2026/03/22: vLLM Router v0.8.0: Prompt Cache, Shared Prefix Routing, and a Dashboard https://barrahome.org/2026/03/22/vllm-router-v080-prompt-cache-and-observability.md - 2026/03/20: vLLM Router v0.7.x: The Road to Enterprise Routing https://barrahome.org/2026/03/20/vllm-router-road-to-enterprise-routing.md - 2026/03/13: NVIDIA NIM vs NVIDIA Dynamo: two different answers to the same problem https://barrahome.org/2026/03/13/nvidia-nim-operator-vs-dynamo.md - 2026/03/06: LMCache-Aware Routing: When Prefill Workers Stop Being Stateless https://barrahome.org/2026/03/06/vllm-router-lmcache-aware-routing.md - 2026/03/05: vLLM Router: the story of a fork and the features upstream doesn't have https://barrahome.org/2026/03/05/vllm-router-fork-production-features.md - 2026/02/19: On RAG Applications: From Theory to Enterprise Setups https://barrahome.org/2026/02/19/rag-from-theory-to-enterprise.md - 2026/02/09: pmap: debugging memory layout and understanding process address space https://barrahome.org/2026/02/09/pmap-debugging-memory.md - 2026/02/08: vLLM KV Offloading: Key Findings from the Official Announcement https://barrahome.org/2026/02/08/vllm-kv-offloading-connector.md - 2026/02/08: LMCache + Redis: Distributed KV Cache for Enterprise LLM Inference https://barrahome.org/2026/02/08/lmcache-redis-distributed-kv-cache.md - 2026/02/07: vLLM router: why prefix-cache-aware routing matters for PD disaggregation https://barrahome.org/2026/02/07/vllm-router-pd-disaggregation.md - 2026/02/06: nginx thread pools: offloading blocking I/O for better performance https://barrahome.org/2026/02/06/nginx-thread-pools.md - 2026/02/06: nginx caching: proxy_cache and fastcgi_cache explained https://barrahome.org/2026/02/06/nginx-caching.md - 2026/02/06: Setting up ngx_markdown_filter_module: a practical guide https://barrahome.org/2026/02/06/nginx-markdown-setup.md - 2026/02/06: A markdown blog with nginx https://barrahome.org/2026/02/06/markdown-blog.md - 2026/02/01: Setting up nginx to serve markdown https://barrahome.org/2026/02/01/nginx-markdown.md - 2025/06/15: Welcome https://barrahome.org/2025/06/15/welcome.md ## Topics & Tags - ai - architecture - blog - bm25 - c - cache - caching - css - debian - debugging - embeddings - enterprise - grafana - inference - kubernetes - kv-cache - linux - llm - lmcache - markdown - memory - meta - monitoring - nginx - nvidia - observability - optimization - performance - python - rag - redis - routing - rust - search - tools - tutorial - vllm ## Notable Features - Custom nginx markdown filter module (author contribution) - Real-time markdown to HTML conversion - Tag-based post filtering - Terminal-inspired UI design - Mermaid diagram support - Open source blog engine ## Source Code GitHub Repository: https://github.com/bet0x/barrahome-2026 ## Content Philosophy Technical depth over breadth. Practical tutorials with real-world examples. Focus on performance, optimization, and understanding how things work under the hood. ## Update Frequency Active blog with regular posts on nginx, performance optimization, and distributed systems. --- Last updated: 2026-03-22