The Iguazio Data Science Platform permits better GPU effectivity and helps to democratize AI infrastructure deployment for each enterprise.

NEW YORK–(BUSINESS WIRE)–May 14, 2020–

VB Transform 2020 Online – July 15-17: Join main AI executives on the AI occasion of the yr. Register today and save 30% off digital entry passes.

Iguazio, the info science platform for real-time machine studying functions, introduced as we speak that the corporate’s resolution for using GPU-as-a-Service has been licensed as a part of the NVIDIA DGX-Ready Software program.

Managing and controlling sources as a part of ML pipelines, which span information processing, mannequin coaching, and inference, can pose challenges reminiscent of inefficient useful resource sharing, workload and information scaling, and excessive growth and operations overhead.

The Iguazio Data Science Platform permits information scientists and information engineers to work in a single self-service information science platform, which dynamically allocates GPUs to workloads for better useful resource effectivity. The resolution helps horizontal scaling, serving to groups handle greater and extra complicated duties extra rapidly, and automates end-to-end ML pipelines from information assortment to mannequin creation. The Iguazio platform is predicated on the main open supply frameworks in a hardened and able to use bundle. It additionally contains the flexibility to observe your entire stack from low-level GPU utilization, by ML jobs and information, to end-to-end pipelines – simplifying the operations of GPU-based techniques.

In addition, Iguazio’s open supply initiatives, Nuclio and MLRun, are the one serverless and orchestration frameworks that handle information and ML workloads, by optimally managing GPU sources, and supporting stateful and distributed workloads, combining simplicity, sooner efficiency, and prices financial savings.

The GPU-as-a-Service resolution from Iguazio is now a part of the NVIDIA DGX-Ready Software program, which delivers confirmed enterprise-grade options that enhance information science productiveness, speed up AI workflows, and enhance accessibility and utilization of AI infrastructure utilizing NVIDIA DGX™ techniques, together with the brand new NVIDIA DGX A100.

Iguazio’s end-to-end information science platform marries efficiency and scale with an open framework that’s deployable wherever – multi-cloud, on-prem or edge.

“Iquazio provides a platform that automates the entire machine learning pipeline from start to finish,” mentioned Yaron Haviv, Co-Founder & CTO at Iguazio. “Certifying the Iguazio Data Science Platform in the DGX-Ready Software program allows users to leverage our optimized AI workflow solution on NVIDIA DGX systems with ease and confidence.”

“The Iguazio Data Science Platform delivers large-scale data processing with both up and down auto-scaling across a pool of shared DGX systems,” mentioned John Barco, senior director of DGX software program product administration at NVIDIA. “This helps users to maximize resource efficiency and accelerate their AI workflows.”

About Iguazio

The Iguazio Data Science Platform permits enterprises to develop, deploy and handle AI functions at scale. With Iguazio, corporations can run AI initiatives in actual time, deploy them wherever; multi-cloud, on-prem or edge, and convey to life their most formidable data-driven methods. Enterprises spanning a variety of verticals, together with monetary companies, manufacturing, telecoms and gaming, use Iguazio to create enterprise influence by a mess of real-time use instances. Iguazio is backed by high monetary and strategic traders together with Samsung, Verizon, Bosch, CME Group, and Dell. The firm is led by serial entrepreneurs and a various crew of innovators within the USA, UK, Singapore and Israel. Find out extra on www.iguazio.com

Iguazio Becomes Certified For NVIDIA DGX-Ready Software Program

View supply model on businesswire.com: https://www.businesswire.com/news/home/20200514005401/en/

Iguazio Press Contact:
Sahar Dolev-Blitental
[email protected]