Element AI Announces New Solution To Help Enterprises Efficiently Manage GPU Clusters

New Element AI Orkestrator product optimizes the use of compute and storage resources by balancing the needs of AI practitioners and IT infrastructure for organizations adopting AI

Montréal, QC (Feb. 11, 2020) – Element AI, a global developer of artificial intelligence-powered (AI) services and software that helps enterprises ‘operationalize’ AI, is pleased to announce Element AI Orkestrator, the latest addition to the company’s portfolio of AI-powered software solutions. Designed for AI practitioners and the IT teams that support them, Element AI Orkestrator is a workload scheduling tool built as a result of the company’s need to optimize its own computing resources. A software-as-a-service (SaaS) version is available now, with an on-premises version expected for Spring 2020.

Element AI Orkestrator helps customers future-proof their IT infrastructures through optimized utilization of GPU clusters, allowing AI practitioners to build quality models and remove most of the engineering heavy lifting required to efficiently use GPUs. By maximizing these specialized and in-demand resources, Orkestrator will play a pivotal role in supporting the structural backbone of organizations deploying AI solutions for increased overall productivity.

“We originally created Element AI Orkestrator for our own internal data science teams and developers conducting AI research. The tool became so useful, our customers began requesting it,” said Ludwig Gamache, Head of IT at Element AI. “Element AI Orkestrator is ideal for companies that need additional computing power to process AI solutions. Our own GPU usage is over 90 per cent; by using the Element AI Orkestrator tool we were able to grow our cluster by 36 times, yet required minimal additions to our IT staff,” added Gamache.

Using Kubernetes, Element AI Orkestrator provides AI researchers, applied research scientists, and developers with the computing infrastructure needed to efficiently build and maintain AI models. Providing enhanced transparency and control over job scheduling, Orkestrator allows AI practitioners to conduct higher-level jobs such as batch processing and hyperparameter searching. IT administrators can simultaneously optimize cluster use, simplify node maintenance, and gain better visibility through dashboards.

“Element AI Orkestrator has a lot of potential for customers who have made significant investments in their AI infrastructures,” said John Barco, senior director of NVIDIA DGX software at NVIDIA. “Element AI Orkestrator can help end-users make the most efficient use and scale of their computing resources,” added Barco.

As organizations build and implement their own AI solutions that will need to be scaled over time—computing power and storage requirements will grow to accommodate heavier GPU-based workloads with increasingly complex resource distribution and management.

To request a sales call or product demonstration for Element AI Orkestrator, contact Element AI.

To learn more about Element AI and its AI-powered products, visit www.elementai.com.

Watch a video demonstration of Element AI Orkestrator:

About Element AI

Element AI is a global developer of AI software that helps people work smarter. Founded in 2016 by serial entrepreneurs including JF Gagné and A.M.Turing Award recipient, Yoshua Bengio, PhD, Element AI turns cutting-edge research and industry expertise into software solutions that exponentially learn and improve. Its end-to-end offering, including advisory services, AI enablement tools and products, aims at helping large organizations operationalize AI and create real business impact. Element AI maintains a strong connection to academia through research collaborations and takes a leadership position in policy-making around the impact of technology on society. For more information, visit www.elementai.com.

© Element AI Inc., 2020, all rights reserved. Element AI™ and the Element AI logo are protected by trademarks of Element AI Inc. Element AI Orkestrator is a trademark of Element AI, and may be registered or pending registration in several jurisdictions. Other trademarks used in this document may be trademarks of the manufacturers or vendors of their respective products.

SOURCE: Element AI

Tags: , , , , ,