Mar 18, 2024

Voltron Data Advances Theseus, Making It the First Petabyte Scale Query Engine for Large Scale Data Processing

Voltron Data

Voltron Data GTC Press

Voltron Data redefining the data analytics SPACE - scale, performance and cost efficiency, with Theseus, the petabyte-scale query engine. Achieving groundbreaking performance, Theseus processes hundreds of terabytes in seconds, redefining speed and efficiency in data-driven decision-making.

SAN FRANCISCO, March 18, 2024 (GLOBE NEWSWIRE) - Today at NVIDIA GTC, Voltron Data, the company accelerating modular and composable data analytics systems, announced that it has significantly advanced spilling out of GPU memory performantly in Theseus, the industry’s first accelerator-native distributed query engine built to leverage the latest hardware and software innovations to solve data challenges too big and time-sensitive for Spark. With this new release, Theseus has achieved a new level of performance and scale, completing the full TPC-H 100 terabyte scale factor on unsorted Parquet files directly from storage in less than an hour with only 5 terabytes of GPU memory.

Hyoun Park, CEO and Chief Analyst of Amalgam Insights, said:

“Voltron Data is playing an important role in bridging analytics and AI infrastructure usage in the Era of AI. As the C-Suite moves forward with hardware investments, they should be aware of the role GPUs can play to accelerate analytic and data driven decision-making by orders of magnitude. As companies face the imperative to incorporate massive data sources into timely decisions, they must take full advantage of modern hardware, data platforms and the most current data.”

Josh Patterson, co-founder and CEO of Voltron Data, said:

“Theseus enables enterprises to use GPUs to quickly analyze log data, machine data and large amounts of tabular data generated on the fly – this is because GPUs are great at processing data, especially for data moving too fast to thoroughly index due to its sheer computational power. For example, cybersecurity and financial systems generate dozens of terabytes of data a day. For some use cases, the time to ingest and process the data as fast as possible is all that matters, people want the freshest data being used always. With Theseus, we can help organizations cut their total time to insights down by orders of magnitude, moving to seconds and minutes by skipping long ingest times and using GPUs to power through data.”

note icon
Patterson will give a talk on Theseus and its newest capabilities in Lambda Labs’ booth at NVIDIA GTC on March 19 and 20 at 2 p.m.

Ingesting petabytes of data for analytics takes enterprises hours to days today, and now queries can run directly on data in a matter of seconds and minutes. This opens a whole new class of use cases for enterprise organizations limited by CPU-based data systems. Theseus points to data where it lies and uses hardware accelerator’s sheer computational power to analyze massive data as quickly as possible.

Theseus is built from the ground up to help companies take advantage of GPUs and full system hardware accelerators such as high bandwidth memory, accelerated networking and storage. Theseus enables data preprocessing for AI/ML workloads on GPUs which unifies data analytics and AI pipelines on the same infrastructure, while lowering vendor lock-in risk, energy consumption and data center footprint. Theseus is available to enterprises and government agencies as well as embedded through partners. HPE is the first partner to embed Theseus as its accelerated data processing engine as part of HPE Ezmeral Unified Analytics Software.

About Voltron Data

Voltron Data offers a new way to design and build composable data systems for enterprises and government agencies. For more on Voltron Data, simply go to or follow @VoltronData.

Media and Analyst Contact

Amber Rowland