Sign in
Sign Up
live scale simulation

PostgreSQL bucklesat 100M rows.

JSONB columns balloon in size. Vacuuming stalls. Index bloat becomes a full-time job. It was built for structured rows — not 850 KB JSON blobs at scale.

Record type:EMR Patient Chart
Raw size:850 KB
Hydrous compressed:~255 KB gzip
Composite indexes:0 — forever
Day 1
100K
500K
1M
5M
10M
1B
1/7
🏥
1 record
Your EMR goes live
HydrousDB
GCS + FIRESTORE HYBRID
0.3ms
latency
monthly cost
$0
composite indexes
0 needed ✓
Zero composite indexes. YYMMDD prefix stamps the ID. Blob compressed to ~255KB on GCS.
Firestore
DOCUMENT STORE
1.2ms
latency
monthly cost
$0
composite indexes
0 needed ✓
Simple document write. No issues at this scale.
Supabase
POSTGRESQL + JSONB
1.8ms
latency
monthly cost
$0
composite indexes
0 needed ✓
Row inserted. JSONB column stored fine.
MongoDB Atlas
DOCUMENT STORE
1.5ms
latency
monthly cost
$0
composite indexes
0 needed ✓
Document stored in Atlas. All good for now.
<1ms
Warm cache read
LRU in-memory
0
Composite indexes
required, ever
70%
Storage savings
via auto gzip
$0.026
Per GB / month
storage cost

Trusted by developers building the next generation of SaaS

500+
Projects deployed
2.4B
Records stored
99.9%
Uptime SLA
Pay-as-you-go
No fixed plans

Built for

Know if HydrousDB fits your system

We're transparent about where we excel — and where we don't.

Multi-tenant SaaS PlatformsEvery customer defines their own fields. No index conflicts, no schema migrations, ever.
// e.g. Airtable, Notion, Monday.com — each workspace has unique columns.

⚡ Bucket-per-tenant isolation + zero composite indexes = infinite schema flexibility.

Form & Survey BuildersUsers design forms at runtime. Fields unknown at build time. Store and query any response structure natively.
// e.g. Typeform, JotForm, Tally — each form is a different schema.

⚡ Any 3 fields become queryable on the fly. The rest lands in the compressed GCS blob.

Internal Tooling PlatformsEach company configures completely different data models. No waiting for indexes to build.
// e.g. Retool, Appsmith, Budibase — customers connect different schemas.

⚡ Any customer, any field, any query — instantly, without ops intervention.

Event & Audit Log StorageHigh-volume append-only streams where event shapes vary by type and producer.
// e.g. Segment-style pipelines, Mixpanel clones, audit trails for SaaS apps.

⚡ Compressed blobs at $0.026/GB vs Firestore $0.17/GB. 1B events: $26 vs $170/mo.

AI & LLM Output StorageLLM outputs vary per prompt. No migrations when your schema changes.
// e.g. LangChain traces, function call outputs, RAG pipeline results, agent memory.

⚡ LLM JSON compresses 80%+. No schema = no friction as your prompts evolve.

No-Code / Low-Code BackendsEvery app has a different data model defined by the end user at runtime.
// e.g. Bubble, WeWeb, FlutterFlow, Glide — user-defined data structures per app.

⚡ Zero index setup + pay-per-use + scales to zero = ideal generated-app backend.

Architecture

Why zero indexes works

Timestamp-encoded doc IDs replace orderBy. No composite index, ever.

01
Timestamp-encoded IDs
Document IDs are zero-padded timestamps + recordId. Firestore ordering by __name__ is identical to created_at. Zero composite indexes regardless of field names.
0001748534400000_rec_01JA2X...
02
Tiny Firestore index, big GCS record
Only 3 queryable fields live in Firestore (~200 bytes/doc). The full record lives compressed in GCS.
{ recordId, gcsPath, field1, field2, field3 }
03
LRU in-memory cache layer
Hot records live in a LRU in-memory keyed by recordId. Warm reads are sub-millisecond with zero network calls.
recordCache.get(id) → <1ms — zero network
04
Auto-compressed blobs
Records over 20KB are gzip-compressed before GCS upload. Storage cost drops 70%. Decompression is fully transparent.
gzip if size > 20KB → ~70% smaller

◉ Live Architecture Tour

The record that
never touches a database

⬇
Phase 1
Ingestion
⚡
Phase 2
Retrieval
🌍
Phase 3
At Scale
🔎
Phase 4
Smart Query
Pipeline
🖥
API
Request in
🗜
Gzip
Compress
🗺
Route
ID→Path
☁
GCS
Upload
hydrousdb — ingest — 800KBLIVE
▊
⬇
Phase 1 · Ingestion

Every medical record starts as raw JSON. We stamp it with a time-encoded ID, shrink it with gzip, and drop it straight to cloud storage — zero database involved.

Architecture Overview
⬇
Phase 1 · Ingestion
A record arrives. Watch it get prepared, compressed, and stored.
⚡
Phase 2 · Retrieval
Fetching a record without ever touching a database.
🌍
Phase 3 · At Scale
One billion records. The math that changes everything.
🔎
Phase 4 · Smart Query
Query an entire month of records. Still zero Firestore reads.

At scale

10 million records.
One database that doesn't flinch.

At 10 million records, architecture matters more than pricing tiers.

$110
HydrousDB cost at 10M records
↓ vs Firestore $960/mo
4ms
Avg query latency (cold path)
↓ vs Firestore 140ms
0
Composite indexes required
↓ Firestore needs 12+ at this scale

How we compare

HydrousDB vs the alternatives

At 10 million records, architecture matters more than pricing tiers.

MetricHydrousDBFirestore
Cost at 10M records / monthScale
$110$960
Composite indexes at 10M recordsIndexes
Zero — ever12+ required
Query latency (cold, 10M records)Latency
<5ms cached80–200ms
Max record sizeRecord Size
Unlimited1 MB
Storage cost per GB / moStorage
$0.026$0.170
Pay only for what you usePricing
✅ Always✅ Yes
Generous forever-freeFree Tier
100k reads, 10k writes, 0.5GB50k reads, 20k writes

Cost intelligence

Your bill vs everyone else's

Real numbers. Real scenarios. Watch HydrousDB stay flat while competitors compound.

Live pricing data

cost trajectory

HydrousDB stays flat — others compound

Real customer growth data

Monthly bill for a SaaS product growing from 0 → 1M users

Pricing

Pay for what you use

No plans, no tiers, no surprises. Generous free allowance resets every month.

Storage

$0.026

/ GB / mo

Reads

$0.25

/ 100K reads

Writes

$1.92

/ 100K writes

Deletes

$1.43

/ 100K deletes

Estimate your bill

Reads / mo500K
Writes / mo100K
Storage (GB)10GB
Est. monthly cost$2.97/mo

First 100,000 reads, 10,000 writes & 0.5GB free every month.