Trailer 60 seconds
What is Prithvi?
Prithvi is an open-source, in-memory key-value database written in Java — built from scratch for performance, transparency, and developer control. It supports multiple data types, expiration, LRU caching, disk persistence, and a multi-threaded architecture.
Built from Scratch
No frameworks. No databases under the hood. Everything handcrafted.
Designed for Control
Get full visibility and tweakability over TTLs, eviction, and internal memory behavior.
Accessible via SDK
Use the official JavaScript SDK to connect, query, and persist from anywhere.
What did I cook?
In-Memory Database
No disk reads. No delays. Just pure RAM-powered speed.
TTL Expiry
Every Time-bound key is removed by an async thread.
LRU Cache
10,000 current keys supported. No bloating. Old keys are evicted.
JSON Persistence
Power goes out. Data stays. Always prepared.
Write-Ahead Log
Every write is logged, crash-safe by design.
Multi-thread handling
Serves concurrent clients on separate threads, no waiting in line.
Rich Collection of Types.
More than strings! Support for lists and sets with native ops.
Prometheus Metrics
Emits real-time internals observability built-in.
Command Protocol
Custom TCP interface. Lightweight, precise, controlled.
Benchmarks
Impressed yet? Support Us
Python SDK
Prithvi offers official SDKs for multiple languages. Choose your preferred environment.
pip install prithvi-py-sdkMinimal CLI
Start and use Prithvi in under 1s. No bloat. No setup.
Zero Dependencies
Written in raw Java. Cross-platform. No installs beyond Java 11.
Wired to GitHub
Full open-source code, docs, and SDK hooks live on GitHub.
Prithvi v Redis
Disclaimer: Prithvi is a prototype key-value store focused on feature completeness and correctness, not raw performance. It runs on the JVM with a basic architecture, built-in garbage collection, and no low-level optimizations. Redis, by contrast, is a mature, production-grade system written in C with performance as a core priority. These benchmarks reflect architectural trade-offs, not parity in design goals.
Note* Benchmarks were conducted on an Apple MacBook Air (M1, 2020) with 16 GB RAM, 256 GB SSD,
7 GPU cores, 8 efficiency CPU cores, and a 16-core Neural Engine. Results may vary on different hardware.
Download Now
Download Prithvi — with TTL, LRU, persistence, and SDK access baked in.
Meet The Developer
I'm an engineer obsessed with building things from first principles. Languages, databases — the closer to the metal, the better.
Prev Works
Ganges — A Sanskrit-inspired programming language. From lexer to VM, all handwritten.
Rooster — The simplest way to deploy your React app on S3.
Finalist at Smart India Hackathon 2023. Built under constraints. 5X hackathon winner.
"Don’t just use systems. Build them. Understand them.