DataCore · Big Data Platform

The lakehouse engine built for exabyte-scale data.

DataCore unifies your data lake and warehouse into a single, cloud-native platform — running batch and streaming workloads on one elastic compute engine. Store everything, query anything, and scale linearly from terabytes to exabytes without re-architecting.

Start Free Trial → View Pricing

Single-namespace storage

10PB+

Daily compute throughput

<500ms

OLAP query latency (p95)

Distributed Cluster 64 nodes · healthy

2.4 TB/s

Sustained ingest

98.7%

Cluster utilization

Failed shards

Unified Architecture

One lakehouse. Five layers. Zero data movement.

DataCore collapses the traditional lake-plus-warehouse stack into a single open-format platform. Open table formats (Apache Iceberg & Delta) give you ACID transactions, time-travel, and schema evolution directly on object storage — no proprietary lock-in.

Serving & OLAP

Vectorized MPP query engine with cost-based optimization and result caching. Sub-second analytics over petabyte tables for BI tools, ad-hoc SQL, and data apps.

Unified Compute

A single engine executes batch ETL, continuous streaming, and federated queries. Decoupled compute pools scale independently per workload with workload isolation.

Open Table Format

Apache Iceberg / Delta tables deliver ACID guarantees, snapshot isolation, time-travel, and in-place schema evolution on commodity object storage.

Catalog & Governance

Unified metadata catalog with column-level lineage, fine-grained access policies, data masking, and a full audit trail for SOC 2, GDPR, and PDPA.

Object Storage

Tiered storage across hot, warm, and cold tiers on S3, GCS, Azure Blob, or on-prem MinIO — with transparent compaction, Z-ordering, and compression.

Capabilities

Everything a platform team needs, built in.

⚡

Stream–Batch Unified Engine

A single execution model for bounded and unbounded data. Write a pipeline once and run it as a nightly batch job or a continuous sub-second stream — no rewrite, no second system.

1 engine

⤢

Elastic Auto-Scaling

Compute pools expand and contract with demand in seconds. Scale-to-zero on idle, burst to thousands of cores under load — you pay only for what runs.

◷

Time-Travel

Query any historical snapshot and roll back bad writes instantly.

🔒

Governed by Default

RBAC, column masking, and full audit logging out of the box.

⛁

Federated Query Fabric

Join across the lakehouse, external warehouses, and live operational databases in one SQL statement — without copying data first.

☁

Multi-Cloud

One control plane across AWS, Azure, GCP & on-prem.

Performance

Benchmarked for production scale.

Representative figures from a 64-node reference cluster running mixed analytical and streaming workloads.

Dimension	DataCore	What it means for you
Max namespace size	Exabyte-scale	One logical platform for every dataset in the enterprise
Sustained ingest	2.4 TB/s peak	Absorb the highest-volume event firehoses without back-pressure
OLAP query latency	< 500 ms (p95)	Interactive dashboards over petabyte fact tables
Streaming latency	Sub-second end-to-end	Real-time fraud, IoT, and operational analytics
Concurrency	10,000+ queries	Serve the whole organization from one cluster
Availability SLA	99.99%	Multi-AZ replication with automatic failover

Pricing

Transparent, per-node licensing.

DataCore is licensed per compute node. Scale your cluster up or down at any time — you are only licensed for the nodes you run.

Starting from

DataCore node license

S$499

per node · billed annually

Stream–batch unified compute engine
Open lakehouse (Iceberg / Delta) storage
Vectorized OLAP & federated query
Catalog, lineage & governance suite
Multi-cloud & hybrid deployment

S$499 is the entry list price per node. Final pricing scales with cluster size, support tier, and deployment model — volume and committed-use discounts apply.

Estimate your cluster

Drag to size your deployment. Most production clusters start at 8–24 nodes.

Compute nodes 20 nodes

3200

Est. list price / year S$9,980

The Platform

Better together.

DataCore is the foundation. Pair it with the rest of the INFINITEMIND stack to feed and activate your lakehouse.

Ready to consolidate your
data stack on DataCore?

Start a free 14-day trial on a managed cluster, or talk to our solutions team about a reference architecture for your workloads.

Start Free Trial → Book a Demo