PUCE Storage · Enterprise

PUCE Storage

Enterprise storage with deduplication, compression, and quantum-safe encryption.

Try it — your file, post-quantum encrypted

Real upload with your account: encrypted in the browser (ML-KEM-768) BEFORE it leaves, presigned URL, never through our server.

Zero-knowledge: only I can open it (the key stays in my browser; not even the service can decrypt).Choose a fileup to 100 MB in this demo

No objects yet. Upload a file above.

Storage redesigned for the post-quantum era

Three pillars that set PUCE Storage apart from any legacy solution.

Smart Deduplication

Eliminates duplicate data before compression, reducing storage costs by up to 60%.

Multi-Cloud

Integrates with AWS S3, Azure Blob, Google Cloud Storage, and on-premises storage.

Compliance

Ready for FIPS 140-3, HIPAA, GDPR, LGPD. Self-assessment tools included.

Storage pipeline

5 stages executed on every write operation — fully transparent to the client.

Ingestion

Data arrives via SDK (Python/Node/Rust/Go) or REST API. Each object is split into 64 MB chunks for parallel transfer.

Deduplication & Compression

Inline deduplication via content hashing. High-ratio lossless compression for optimal CPU–space balance. Typical 60–75% reduction in stored volume.

PQ Encryption

Each chunk is encrypted with ML-KEM-768 + AES-256-GCM (FIPS 203). Ephemeral keys never persisted. Optional client-side encryption for zero-knowledge.

Automatic tiering

Hot → Warm → Cold based on access frequency. Transparent migration without URL changes. Configurable retention per bucket (7d – 10 years).

Multi-region replication

Synchronous or asynchronous replication across 3+ regions. Automatic failover. Compliance with data residency requirements (GDPR, NIS2).

Use cases

Four scenarios where PUCE Storage is the obvious choice.

Enterprise backup

Replace legacy backup solutions with PQ-encrypted storage. Deduplication cuts volume by 60-75%. Automatic retention and tiering eliminates manual management.

Regulated data

Healthcare (HIPAA), financial (PCI-DSS) and government (NIS2) data requires strong encryption and audit trail. PUCE Storage provides both by design.

Post-quantum data lake

Store petabytes of analytics data with PQ protection from ingestion. S3-compatible API integrates with Spark, Trino and existing tools.

Legal & forensic archive

Legal documents and forensic evidence need long-term verifiable integrity. SHA3-256 hash + Merkle tree provides tamper-proof evidence.

PUCE Storage vs traditional storage

Six critical differences the auditor will check.

Feature	PUCE Storage	Traditional storage
Encryption	ML-KEM-768 + AES-256-GCM	AES-256 (vulnerable to Grover)
Quantum resistance	ML-KEM-768 (FIPS 203)	None
Deduplication	Inline chunk-level	Post-process (if available)
Tiering	Automatic Hot/Warm/Cold	Manual or lifecycle rules
Integrity	SHA3-256 + Merkle tree	MD5/SHA-256 checksum
Multi-region	Sync/async with PQ-TLS	TLS 1.2/1.3 (no PQ)

Backup software integration scenarios

4 ways to integrate PUCE with any backup system — each with different trade-offs between deduplication, compression and post-quantum protection.

Scenario A — Compress-Only Pre-Backup

Recommended · Universal

App/DB (100 TB) → PUCE compress-only (30 TB) → Backup dedup+compress → ~15 TB

Typical ratio:6-7:1

Backup dedup works (data is not encrypted)

Compatible with any backup software

Reduces FETB licenses — pay 30 TB instead of 100 TB

⚠No post-quantum protection (encryption is classic from backup)

Scenario B — Full Pipeline Pre-Backup

Dedup caveat

App/DB (100 TB) → PUCE compress+encrypt+sign (~22 TB) → Backup (dedup ~0%) → ~22 TB

Typical ratio:4-5:1

Full PQ end-to-end protection

Compatible with any backup software

⚠Backup dedup does not work (encrypted data looks random)

⚠Only recommended when dedup is not needed

Scenario C — Integrated in Backup Engine

Deep integration

App/DB (100 TB) → Backup [Dedup → PUCE Engine (replaces compression) → PQ Encrypt] → ~10 TB

Typical ratio:8-10:1

Dedup works 100% (PUCE acts after)

ML-select compression superior to native

PQ encryption on storage

⚠Requires deep integration (SBT_TAPE for Oracle, VDI for SQL Server, etc.)

⚠Support varies by backup system

Scenario D — Post-Dedup, Pre-Storage

Ideal · Best overall

App/DB (100 TB) → Backup [Dedup → 40 TB] → PUCE Storage [ML-Select Compress → PQ Encrypt+Sign] → ~8 TB

Typical ratio:10-15:1

Dedup works 100%

ML-select compression on already deduped data

Full PQ encryption on storage

Digital signature on each block

Compatible with all systems

⚠Does not reduce FETB license (backup sees original volume)

⚠Requires access to post-dedup point

Scenario	Dedup works?	PQ Encrypt?	Integration	Ratio (100 TB)	Best for
A · Compress-Only	✓ Sim	✗ Não	None	6-7:1	Reduce FETB licenses
B · Full Pipeline	✗ ~0%	✓ Sim	None	4-5:1	PQ end-to-end, no dedup
C · Motor Integrado	✓ Sim	✓ Sim	Deep (API)	8-10:1	Oracle RMAN, SQL Server
D · Pós-Dedup	✓ Sim	✓ Sim	Storage tier	10-15:1	Best possible ratio

General Comparison — All Platforms

Integration method, recommended scenario and ratios for each backup system.

Platform	PUCE integration method	Recommended scenario	Compression ratio (without PUCE → with PUCE)	Dedup works?	PQ Encrypt?	Complexity
Oracle RMAN	SBT_TAPE library (libpuce_sbt.so)	C (Integrated Engine)	2.1:1 → 9.3:1	✓ Excellent (block-aligned)	✓	Medium
Commvault	CVPySDK + Cloud Library S3	A+D (Compress + Archive)	2.5:1 → 10-15:1	✓ MSDP + PUCE	✓	Low-Medium
Veritas NetBackup	Hook scripts (bpstart/bpend_notify) + OST	D (Post-Dedup)	2.5:1 → 10-15:1	✓ MSDP + PUCE	✓	Low-Medium
Dell NetWorker	Cloud Boost S3 / AFTD / Pre-save scripts	D (Data Domain + PUCE archive)	2.0:1 → 10-14:1	✓ DD + PUCE	✓	Medium
Veeam B&R	S3 Object Storage Repository (native)	A+D (best cost-benefit)	1.5:1 → 10-15:1	✓ MSDP + PUCE	✓	Low (plug & play)
SQL Server	VDI / BACKUP TO URL (S3) / .bak → PUCE	C (VDI) or A (BACKUP TO URL)	1.8:1 → 7-10:1	✓ (8K pages repeat)	✓	Medium
PostgreSQL	pgBackRest S3 / pg_basebackup pipe / WAL archive	C (pgBackRest S3)	2.3:1 → 6-9:1	✓ (WAL dedup excellent)	✓	Low