Teradata Brings Elastic Python Compute to AI Studio, Unifying Data, Governance, and AI at Enterprise Scale

Local model hosting has no external API dependencies or token-based inference costs.

Rhea-AI Summary

Teradata (TDC) announced general availability of Elastic Python Compute within AI Studio, providing a managed environment for distributed AI workloads. Built on Ray, a framework that distributes computing tasks, the capability supports parallel Python execution for data engineering, model training and custom applications without manual cluster management.

Workloads run within the Teradata platform, with direct access to its database, S3 storage and supported open-table catalogs. The capability is available on AWS and Microsoft Azure, and on-premises through Teradata Factory. Teradata expects Google Cloud availability in Q4 2026.

Key Figures

Arrow Flight SQL transfer speed: up to 84x faster Credit scoring pipeline runtime: three times faster Feature engineering runtime: from four hours to twenty minutes +2 more

Arrow Flight SQL transfer speed

up to 84x faster

Teradata benchmark versus traditional row-based transfer methods

Credit scoring pipeline runtime

three times faster

Customer example using in-database analytics

Feature engineering runtime

from four hours to twenty minutes

Telecommunications provider customer example

Model accuracy improvement

12%

Telecommunications provider example; trained on the full dataset

Expected Google Cloud availability

Q4 2026

Elastic Python Compute

Previous AI Reports

1 past event · Latest: Sep 02

Same Type 1 event

1. Sep 02

   AI data integration

   24h Move

   -1.4%

   Integrated the platform with OneLake to run AI on stored data without migration.

24h Move is the share-price change in the day after each event; other market factors may also have contributed.

Key Terms

arpu, gdpr, massive parallel processing

3 terms

arpu financial

"call drop rates, signal quality indices, and ARPU (average revenue per user) trends"

ARPU, or Average Revenue Per User, measures how much money a company earns, on average, from each of its customers over a set period. It helps investors understand how effectively a business is generating income from its customer base, similar to calculating how much each customer spends at a store. Higher ARPU often indicates stronger sales per customer and better revenue performance.

gdpr regulatory

"satisfying both GDPR obligations and national data residency requirements"

General Data Protection Regulation is a law that sets rules for how organizations must collect, store and use personal data about people, and gives individuals rights over that data. It matters to investors because noncompliance can lead to large fines, higher operating costs and damaged reputation, while strong compliance can be a competitive advantage—think of it as a strict safety code for handling customer information.

massive parallel processing technical

"Massive parallel processing supports large-scale AI and data workloads"

Massive parallel processing is a computing approach where thousands or millions of small processors or cores work at the same time to solve parts of a problem simultaneously, like a large team of workers each handling one piece of an assembly line. For investors, it matters because it enables much faster analysis of large datasets, real-time services, or advanced AI models, which can drive product capability, scale, operating cost and competitive advantage.

AI-generated analysis. How Rhea-AI works. Not financial advice.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

New distributed Python capability accelerates large-scale AI operations while keeping data and compute within the Teradata platform

SAN DIEGO, Oct. 14, 2026 /PRNewswire/ -- Teradata (NYSE: TDC) today announced the general availability of Elastic Python Compute within AI Studio. The capability gives enterprise data science teams a managed, distributed Python environment for scaling AI compute and machine learning workloads through parallel Python execution for data engineering, machine learning, and model development.

Elastic Python Compute is built on Ray, a distributed computing framework, and integrates natively with the Teradata Autonomous Knowledge Platform and AI Studio. Data scientists can parallelize model training, execute custom Python applications at scale, and serve high-scale ML and AI workloads without the overhead of manual cluster setup or managing Ray clusters. Elastic Python Compute supports both single-server and cluster deployments, giving organizations the flexibility to scale from individual workloads to enterprise-scale operations, with faster development cycles, stronger governance controls, and lower infrastructure costs.

Keeping Data and AI Compute Within the Platform
Elastic Python Compute supports local model hosting and governance, with no external API dependencies and no token-based inference costs. AI workloads run entirely within the Teradata platform, including AI Studio, with no data movement required. Elastic Python Compute reads and writes data directly to the Teradata Database, S3 object storage, and Open Table Format catalogs, such as Apache Iceberg and Delta Lake, enabling data science teams to incorporate files and images for model training and scoring. Flexible cluster management, including the ability to run separate clusters for different workloads, allows teams to optimize cost and performance by selecting CPU or GPU compute based on the task at hand. Massive parallel processing supports large-scale AI and data workloads across cloud, on-premises, and hybrid deployments.

The platform also supports Arrow Flight SQL, enabling parallel, columnar data transfer between applications and the Teradata platform at speeds up to 84x faster than traditional row-based transfer methods, based on Teradata benchmark testing. *

Customer Impact
A retail bank had been exporting tens of millions of customer records to local Python environments for credit feature engineering, a practice that created data governance gaps and put the organization at risk of breaching data residency regulations. With AI Studio's in-database analytics, credit scoring pipelines covering payment history, utilization ratios, and delinquency flags now run three times faster while remaining within the governance perimeter. A complete audit trail from raw data through feature engineering to final model output allows regulators to replay any step in the process on demand.

For a major telecommunications provider, the challenge was scale. Its data science team had been exporting billions of call detail records and network telemetry events to external Python clusters, a slow, costly process that created data sovereignty risk. Feature engineering across call drop rates, signal quality indices, and ARPU (average revenue per user) trends now runs directly inside the Teradata platform. The churn prediction model trains entirely within the provider's own infrastructure, satisfying both GDPR obligations and national data residency requirements. Feature engineering runtime fell from four hours to twenty minutes, and model accuracy improved by twelve percent as a result of training on the full dataset rather than a constrained sample.

In the public sector, a government maritime agency needed to track vessels switching off their transponders to obscure their movements. Using Elastic Python Compute, the agency's data science team combined satellite imagery, radar signals, AIS data, and commercial imaging to reconstruct vessel movements that would otherwise go undetected. AI-driven analytics fused these diverse data sources into actionable maritime intelligence, improving visibility and enabling faster, more effective responses to security threats at sea.

Executive Perspective 
"Enterprise data science teams have spent years working around the fundamental constraint of moving data outside their platforms to build and train AI models, accepting the governance risk and cost that comes with it. Elastic Python Compute removes that constraint. Teams can now run distributed AI workloads at scale where their data already lives, and get to production faster without compromising on governance or cost control."
— Sumeet Arora, Chief Product Officer, Teradata

Availability
Elastic Python Compute is generally available within AI Studio on AWS and Microsoft Azure. Availability on Google Cloud is expected in Q4 2026, with on-premises available now with Teradata Factory.

* Performance results may vary. Benchmarks were conducted in a specific test environment and may not reflect results in your infrastructure.

About Teradata
Teradata empowers enterprises to turn intelligence into autonomous action, grounding AI agents in deep business context and trusted data. As AI agents multiply, Teradata is the context foundation, governance layer, and performance backbone that companies need now. The Teradata Autonomous Knowledge Platform puts AI into production across cloud, on-premises, and hybrid environments.

The Teradata logo is a trademark, and Teradata is a registered trademark of Teradata Corporation and/or its affiliates in the U.S. and worldwide.

MEDIA CONTACT
January Machold
January.Machold@Teradata.com

View original content:https://www.prnewswire.com/news-releases/teradata-brings-elastic-python-compute-to-ai-studio-unifying-data-governance-and-ai-at-enterprise-scale-302908570.html

SOURCE Teradata Corporation

FAQ

AI-generated questions and answers. How Rhea-AI works. Not financial advice.

Where is Teradata Elastic Python Compute available?

Elastic Python Compute is generally available within AI Studio on AWS and Microsoft Azure, with on-premises availability through Teradata Factory. Teradata expects availability on Google Cloud in Q4 2026.

What deployment and compute options does Teradata Elastic Python Compute support?

Elastic Python Compute supports single-server and cluster deployments, with CPU or GPU compute selected for individual tasks. Teams can run separate clusters for different workloads, and the capability supports cloud, on-premises and hybrid deployments.