Use Case

VeloDB for AI

The AI-Ready analytics database for the AI era.

Real-Time Analytics for AI

From Customer-Facing Analytics to Agent-Facing Analytics

Customer-Facing Analytics

As cloud computing and SaaS software become more popular, embedding analytics into applications is crucial. This is also called custom-facing or user-facing anlytics.

Order Analytics

Advertising Analytics

Inventory Analytics

Agent-Facing Analytics

With the rise of AI technologies, especially AI Agents, more analytical decisions will be made automatically by AI. This will improve efficiency and accuracy in decision-making.

Fraud Detection

Ad Serving

Personalized Recommendation

VeloDB is Ready for Agent-Facing Analytics

~ 1 s

minimum data latency

Real-Time Ingestion & Update

< 100 ms

average query latency

Blazing-Fast Analytics

> 10,000 QPS

maximum query concurrency

High-Concurrent Queries

MCP Server

Seamlessly Integrated with AI Agent

Lakehouse Analytics for AI

Lakehouse: The AI-era data infrastructure unifying analytics and machine learning

Analytical Workloads

Long-running ETL

Machine Learning

Lightweight ETL

Interactive Analytics

Open Data Lakehouse

Lakehouse Compute

Batch Processing Engine

(Spark, ...)

Real-Time Analytics Engine

(VeloDB)

Lakehouse Storage

Data Lake

(Iceberg, Hudi, Delta Lake, ... )

Catalog

(Polaris, Unity, Glue, ...)

Data Sources

Tables

Streams

Files

...

Real-Time Analytics Engine

Use VeloDB as the real-time analytics engine, primarily responsible for supporting interactive analytics and lightweight ETL computational workloads.

Batch Processing Engine

Use Spark-like batch processing engines, primarily responsible for supporting long-running ETL and machine learning computational workloads.

Open Lakehouse Storage

Build an open lakehouse storage based on Data Lake using open table formats and open Catalog.

VeloDB is Ready for AI Lakehouse

Data Preparation

Use the fastest SQL analytics engine to filter, sample, and prepare datasets from massive amounts of data for model training and inference.

Feature Engineering

Use the fastest SQL analytics engine to perform data cleaning and preprocessing and feature extraction and transformation.

Quality Evaluation

Leveraging the fastest SQL analytics engine, we perform rapid, multi-dimensional analysis on quality data from both test and online environments.

Observability for AI

The Two Major Drivers of Observability's Evolution

Cloud-Native & Micro Services

This era introduced complex distributed systems, demanding unified observability for countless services with logs, metrics, and traces.

AI-Native & AI Agents

AI Observability is crucial in the Agent era to manage the exponential complexity and data volume from autonomous agents and LLMs.

VeloDB is Ready for AI Observability

Massive Data Volume & Cost Efficiency

VeloDB efficiently handles huge AI data volumes, significantly cutting storage costs with advanced compression and smart data tiering.

High Throughput Ingestion & Fast Queries

VeloDB offers high-throughput ingestion up to GB/s and achieves sub-second real-time search for AI contextual text data in log and trace.

Seamless LLM Ecosystem Integration (Coming soon)

VeloDB integrates out-of-the-box with key LLM tools like Langfuse and LangSmith, simplifying AI application monitoring and optimization.

Unstructured Data Analytics for AI

Hybrid Search for RAG (Coming soon)

VeloDB supports Hybrid Search for RAG by combining efficient full-text search with high-performance vector search.

This allows for more accurate and relevant context provision to LLMs, improving generation quality.

AI-Powered SQL in VeloDB (Coming soon)

VeloDB's AI-Powered SQL embeds large language model capabilities directly into SQL functions, enabling powerful semantic text analysis. Analyze textual data with familiar SQL for tasks like sentiment analysis and summarization.

VeloDB for AI

Apache Doris

Company

Partners

Follow Us