Native MongoDB 8.0 Wire Protocol (Port 27017)

MongoDB Flexibility.
Columnar Analytical Power.

Connect Mongoose, PyMongo, or Compass directly on port 27017. Run lightning-fast aggregation pipelines without complex ETL pipelines to separate analytical warehouses.

Port 27017
Native BSON OP_MSG
Full Pipelines
$match, $group, $lookup
Geospatial
$geoWithin, $box, $polygon
Zero Egress
Cloudflare R2 storage
100% Wire Compatible

Works with Your Favorite Document Tools

No code alterations needed. Switch your MONGO_URI to Kolmos and retain all driver logic.

import mongoose from 'mongoose';

// Connect using standard MongoDB URI on Port 27017
await mongoose.connect('mongodb://user:password@cloud.kolmos.dev:27017/analytics_db', {
  tls: true
});

const userSchema = new mongoose.Schema({
  username: String,
  email: String,
  metadata: Object,
  createdAt: { type: Date, default: Date.now }
});

const User = mongoose.model('User', userSchema);

// Complex Aggregation Pipeline executed via DataFusion Arrow engine
const stats = await User.aggregate([
  { $match: { "metadata.tier": "enterprise" } },
  { $group: { _id: "$metadata.region", count: { $sum: 1 } } },
  { $sort: { count: -1 } }
]);
Document Architecture

MongoDB Compatibility Specifications

High-throughput BSON ingestion with vectorized columnar query execution.

MongoDB 8.0 Wire Protocol

BSON OP_MSG on 27017

Native binary framing, SASL SCRAM-SHA-256 auth, and TLS encryption compatible with all official MongoDB drivers.

Aggregation Pipeline

Full Stage Coverage

$match, $group, $project, $sort, $unwind, $lookup, $facet, $bucket, and statistical accumulators.

Geospatial Queries

2dsphere & Legacy

Native support for $geoWithin, $box, $center, $polygon, and $centerSphere spatial bounding filters.

Text & Expression Search

$text & $where JS

Text search matching and JavaScript conditional expression evaluation for advanced filtering.

Unified HTAP Architecture

No ETL Required

Store dynamic JSON documents directly while querying them as high-performance columnar Apache Arrow batches.

Cost Advantage over Atlas

10× to 20× Lower

Stores data on Cloudflare R2 ($0.015/GB/mo) with zero data egress charges, bypassing Atlas IOPS pricing.

Frequently Asked Questions

Can I connect Mongoose and MongoDB Compass directly to Kolmos?

Yes. Kolmos provides 100% wire compatibility with MongoDB 8.0 on standard Port 27017. Mongoose, PyMongo, Compass, and Motor connect using standard mongodb:// connection URIs without any plugin or code modification.

Why use Kolmos instead of MongoDB Atlas?

MongoDB Atlas stores uncompressed BSON on costly cloud EBS volumes and charges high fees for provisioned IOPS and data egress. Kolmos compresses BSON documents into columnar KSF format and stores them on Cloudflare R2 ($0.015/GB/mo) with zero data transfer fees, dramatically reducing infrastructure bills.

How does Kolmos execute fast aggregations over JSON documents?

Instead of scanning raw JSON documents one-by-one, Kolmos translates BSON fields into columnar Apache Arrow memory structures. Embedded Apache DataFusion runs vectorized aggregations ($group, $match, $sort) at multi-gigabyte-per-second memory bandwidth.