Skip to content

NVIDIA Data Processing Solutions

Original document

VERSION 20261022_rev01 · CAPTURED 2026-10-22

This is our stored copy, shown for reference. The original link above is authoritative.

Data Processing Solutions

Data engines ready for AI.

Overview

The Data Foundation for Agentic AI

To transform your enterprise, AI agents need continuous access to your data, putting strain on data infrastructure not designed for agentic reasoning loops.

With NVIDIA infrastructure, enterprises can meet the new volume and velocity of data demands from AI while leveraging their existing data stack. For CPU-based data processing, including analytics, real-time streaming, batch, and online transaction processing (OLTP), NVIDIA Vera delivers the per-core performance and high-bandwidth memory that agentic workloads need. For the fastest speed at scale, NVIDIA CUDA-X™ libraries accelerate the most popular data engines on GPUs, providing agents with on-demand access to structured data in tables and unstructured data living as PDFs, emails, images, and videos across the enterprise.

Accelerating Data Processing for AI Agents

Watch the GTC Berlin keynote to explore how NVIDIA's latest advances in accelerated computing, data processing, and retrieval help enterprises put their data to work for AI agents.

Power Data Processing and Retrieval With NVIDIA Vera CPU

Explore how European cloud providers and system builders are adopting the NVIDIA Vera CPU to support data processing, retrieval, and AI agent workloads.

Benefits

Transform Your Data for AI

Massive Performance Gains

The accelerated computing platform delivers up to a 20x GPU speedup for data processing, while NVIDIA Vera brings a high-performance CPU platform, enabling enterprises to take action faster with new use cases.

Significant Cost Savings

By running on the NVIDIA optimized stack, organizations have saved 80% in costs or more, helping your data infrastructure do more with less.

Easy to Adopt

The world's most popular analytics and vector data engines have drop-in accelerators to make adoption straightforward, including Apache Spark, OpenSearch, and more.

AI-Ready Data

With context from 90% of enterprise data stored in PDFs, messages, and emails with NVIDIA cuVS, and ground truth from terabytes of structured data processed in minutes with NVIDIA cuDF, your data is ready for agentic AI.

Products

NVIDIA Accelerates Data Processing

NVIDIA libraries, tools, and models power the data processing ecosystem with GPU acceleration, delivering faster data workloads and pipelines at enterprise scale.

cuDF for Structured Data

  • Accelerates analytics engines on NVIDIA GPUs
  • Includes drop-in accelerators for Apache Spark, Presto, Polars, and DuckDB
  • Executes analytical queries in minutes from hours

cuVS for Unstructured Data

  • GPU-accelerated vector search and index building for RAG and AI pipelines
  • Integrates with OpenSearch, Elastic, Milvus, and more
  • Reduces vector index build times from hours to minutes

NVIDIA Kumo Structured

  • Leaderboard-winning structured data models for tabular, relational, and time series data
  • Forecasts outcomes, scores risk, detects anomalies, and generates recommendations with a single foundation model
  • Cuts time to prediction from months to seconds, without feature engineering or task-specific machine learning pipelines

NVIDIA Nemotron Retriever

  • Open, leaderboard-winning models that set a new accuracy-efficiency frontier, giving agents the right context with fewer tokens
  • Parses, embeds, and reranks text, images, and complex documents, including PDFs, tables, and charts
  • Lowers agentic cost by 40 percent, reduces incorrect answers by 24 percent, and increases throughput by 2.4x

NVIDIA Auto Ontology

  • Open-source library that generates a starter enterprise ontology from databases, documentation, metadata, and query history
  • Enables business stewards to test, refine, and certify business meaning, giving agents governed context and provenance
  • Delivers up to 90 percent faster ontology creation, 80 percent less time connecting data, and 10x more accurate AI queries

Adopters

Data Processing Ecosystem

From analytical SQL queries to vector search, organizations are adopting NVIDIA's accelerated computing platform into their existing data platforms to accelerate AI-ready pipelines.

Resources

The Latest in Data Processing

  1. Blogs

  2. Sessions

  3. Videos

  4. Hardware Datasheets

Power Data Processing and Retrieval With NVIDIA Vera CPU

Explore how European cloud providers and system builders are adopting NVIDIA Vera CPU to support data processing, retrieval, and AI agent workloads.

NVIDIA cuDF and cuVS Adopted by World's Leading Data Platforms

NVIDIA's accelerated computing platform is fueling modern enterprise data processing. Integrated with the world's most widely used open source data engines-downloaded over 200 million times monthly by developers-these libraries are harnessed across enterprise data platforms, databases, and data lakes.

How Snap Scaled A/B Testing With NVIDIA cuDF

Snap processes 10+ petabytes daily for A/B testing across 940M+ users. Accelerating Apache Spark with NVIDIA cuDF on Google Cloud delivered 4x faster runtimes and 76% cost savings.

Accelerating Large-Scale Analytics With Velox and NVIDIA cuDF

IBM and NVIDIA integrate cuDF with the Velox execution engine, enabling GPU-native query execution for Presto and Apache Spark-delivering up to 12x faster analytics than CPU-only systems.

Accelerating Data Processing for AI Agents

Watch Jensen Huang describe how the NVIDIA Vera CPU delivers higher performance and efficiency to run 3x faster SQL analytics with Starburst and increase the reliability of real-time data streaming with Redpanda by up to 6x.

IBM Reinvents Data Processing

IBM watsonx.data SQL analytics engine Presto is accelerated by cuDF for 5x speedup and 83% cost savings.

Processing 100 Million Rows of Data in Under 2 Seconds With Polars

Polars GPU Engine executes polars code on GPUs for massive speedups.

Data Processing on NVIDIA Vera

With NVIDIA-designed Olympus cores, 1.2 TB/s of memory bandwidth, and a high-speed on-chip fabric, NVIDIA Vera gives agentic AI the per-core performance, data throughput, and predictable performance under load required for data processing workloads, with up to 3x faster SQL analytics and 6x lower p99 latency in real-time data streaming workloads.

Coming soon.

Data Processing on NVIDIA RTX PRO 6000 Blackwell Server Edition

NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs are engineered for intensive data processing and AI workloads, delivering up to 42X speed ups and eliminating pipeline bottlenecks for data at terabyte scale.

Petabyte-Scale Data Processing on NVIDIA RTX PRO 4500 Blackwell Server Edition

NVIDIA RTX PRO™ 4500 Blackwell Server Edition accelerates enterprise-scale data processing with NVIDIA cuDF, delivering up to 8x faster Apache Spark queries than CPU-only systems for AI and agentic applications.

This enables universal acceleration from the data center to the edge.

Next Steps

Ready to Learn More?

Get the latest on data processing news, content, and events.

cuDF

Open source toolkit for structured data using GPU parallelism and memory bandwidth to accelerate data processing and analytics workflows.

cuVS

Open source library for unstructured vector search and data clustering that enables faster vector searches and index builds.

Sign up to receive data science news

Back to the change history