Previous Case Study • AI / Developer tools
CodeDog
AI-assisted security reviews directly in developer pull requests.
Case CDRead full study →
Back to projects
Infrastructure
VedDB
High-performance key-value database designed for operational clarity.
0.2ms
Read Latency
850k/s
Throughput
< 1.5%
Visual Overhead
veddb.velnix.studio
Hit Rate98.42%
Memory Use1.24 GB / 8 GB
$ veddb-cli --monitor
Connecting to ws://localhost:4040/stats...
16:47:01 [CMD] SET usr_9821 → OK (0.12ms)
16:47:02 [CACHE] GET usr_9821 → HIT (0.01ms)
16:47:04 [CMD] LPUSH pending_jobs 8802 → OK (0.19ms)
Overview
VedDB is a lightweight in-memory key-value database written in Go. Built for speed, its core design focuses on visibility, exposing database hit rates, command flows, and memory usage through a real-time terminal interface and reactive web dashboard.
The Stack
GoReactTailwindCSSWebSocketsgRPCeBPF Profiling
What We Shipped
Low-latency Go key-value core
High-frequency WebSocket instrumentation API
eBPF-driven memory profiling metrics
Real-time diagnostic web interface
The Challenge
In-memory stores operate as black boxes. When memory spikes occur or key distribution grows lopsided, developers have to parse system logs or run heavy diagnostic tools in production, which risks causing downtime.
Our Solution
We engineered an ultra-fast key-value core in Go. We embedded low-overhead metrics tracking directly into the command parser and exposed it via WebSockets. We then built a real-time command dashboard showing distribution stats, active keys, and live profiling logs.
Keep Exploring
Other Case Studies
Next Case Study • Mobile / SMB
BizNest
Unified workspace app consolidating invoicing, chats, and scheduling.
Case BNRead full study →