- Products
- SolutionsUse Cases
Real-Time Analytics
Sub-second dashboards and data products on petabytes of data at any concurrency
Data Warehousing
Sub-second analytics on open lakehouse formats with no vendor lock-in
Observability in the AI Era
The most cost-effective alternative to Elasticsearch observability
Context Engineering
Hybrid search and fresh context for RAG, agents, and LLMs
IndustriesFinancial Services
Real-time analytics and fraud detection for banking, capital markets, and on-chain data
Ad & Media tech
Advertiser-facing reporting, content search, and game telemetry on one engine
Automobile & Transportation
Fleet telemetry, real-time recommendations, and multimodal autonomous driving search
- Docs
- Resources
- Pricing
- Contact us
Built for real-time data. Proven in the industries that move fastest.
Prices, campaigns, and telemetry change by the second, and the analytics built on them are often sold to customers. Each industry below shows the workloads teams actually run on VeloDB and what they measured after consolidating.
Financial Services & Web3
Market analytics, payment reporting, fraud and risk monitoring, digital identity, and on-chain data. Prices and positions change by the second, and the analytics built on them are often sold to the customer.
- ~1,000 QPS
- Sustained on 10TB+
- P99 < 1s
- At peak
- 50%+
- Lower operating cost
Ad & Media tech
Advertiser-facing campaign reporting, creative and content search, audience segmentation, and game telemetry. Full-text search and aggregation on the same trillion-row tables, instead of one engine for each.
- 64–90%
- Lower query latency
- 3M rows/s
- Peak write per node
- Under 5%
- Slow queries, was 35%
Automobile & Transportation
Connected-vehicle telemetry, manufacturing traceability, customer profiles and segmentation, live ride-hailing recommendations, and multimodal search over autonomous driving datasets.
- 20x
- Faster segmentation
- ~10 ms
- Single-user profile lookup
- +15%
- Marketing conversion rate
Six more sectors running on the same engine
These do not have a landing page yet, so each one links straight to the customer's published story.
E-commerce & Retail
JD.com runs live A/B testing and conversion-funnel analytics on 10 billion rows a day at 10,000 QPS, after evaluating Druid, Elasticsearch, and ClickHouse.
Read the JD.com storyLogistics & Supply Chain
SF Technology replaced Presto across one million daily queries, cutting P95 latency by nearly 70% and hardware cost by 48%.
Read the SF Technology storyConsumer Electronics & IoT
Xiaomi consolidated Presto, Druid, and Spark into one Doris engine over Paimon, taking aggregation queries from 40 seconds to 8 and concurrency from 5 sessions to 80.
Read the Xiaomi storyGaming
The NetEase Games data team behind Naraka: Bladepoint cut six systems to one on a Doris and Iceberg lakehouse, serving 15 million queries a day.
Read the NetEase Games storyAI & LLM Platforms
MiniMax, the company behind Talkie, moved off Loki and now searches 1 billion log records in under 2 seconds at 10 GB/s write throughput.
Read the MiniMax storyMusic & Entertainment
Tencent Music unified content search and analytics off Elasticsearch, cutting operating cost by up to 80% and storage by 72%.
Read the Tencent Music storyTrusted by teams that work with data at scale
Why these teams consolidated
VeloDB is the real-time analytics and search database from the creators of Apache Doris, available as a managed cloud service or self-hosted.
Fast on the queries production actually runs
Most engines benchmark single-table scans. Real workloads hit multi-table joins on data that is being updated. VeloDB stays sub-second there, which is why teams stop pre-aggregating.
One engine instead of four
Search, analytics, lakehouse queries, and vector retrieval share one copy of the data and one SQL dialect. Your BI tools, dbt, and drivers see MySQL, so nothing in the stack has to be relearned.
Cost that drops when you consolidate
High compression and tiered storage cut the bill, and removing a duplicate copy of the data cuts it again. The reductions on this page are what customers measured after migrating, not list-price comparisons.
Not seeing your industry?
The same engine handles high concurrency, heavy updates, complex joins, and search wherever the data comes from. Talk to an engineer about your workload.
