AI accelerator with 40% better price-performance can run workloads on-demand by being available on the MetaCloud Marketplace.
Santa Clara, CA – cnvrg.io, the operating system for artificial intelligence (AI) and machine learning (ML) built by data scientists, today announced that Habana’s Gaudi AI Accelerator is available on cnvrg.io’s Metacloud marketplace of AI infrastructure solutions. cnvrg.io Metacloud enables AI developers to run AI/ML workloads on a mix of infrastructure and hardware choices, even within the same workflow or pipeline. Today, AI developers can reduce the cost to train models, and innovate faster by optimizing training capacity, effectively merging their on-premises capacity with Amazon EC2 DL1 instances powered by Gaudi accelerators.
Many AI projects are stalled due to the limitations of existing IT infrastructure to meet the growing demands of AI workloads. From the Metacloud marketplace, developers can now choose the Gaudi-accelerated Amazon EC2 DL1.24 xlarge instance for any AI workload, which is optimized for training neural networks. Gaudi-accelerated instances yield up to 40% better price-performance than GPU-based instances.
This latest integration of Gaudi accelerator solution on cnvrg.io Metacloud Marketplace enables AI developers using cnvrg.io Metacloud to take advantage of Habana’s cutting-edge technology on the fly, free from vendor lock-in and the need to develop a new AI stack.
“cnvrg.io and Habana are helping Seagate seamlessly employ our on-premises data center infrastructure and burst incremental capacity onto the AWS’s Amazon EC2 DL1 instance cloud based on Gaudi AI training processors—within a single Seagate AI workflow pipeline,” said Bruce King, Data Science Technologist at Seagate Technology. “Merging our data center resources with AWS EC2 instances via cnvrg.io integration of Habana Gaudi is expected to yield unprecedented productivity results for Seagate and enable us to lower TCO, while accelerating our time-to-train functions.”
“cnvrg.io and Habana are pioneering the AI industry with the combination of two industry firsts: processing power for AI on demand from combined cloud and on-premises infrastructure and AI training instances by AWS based on Habana Gaudi,” commented Yochay Ettun, CEO and Co-Founder at cnvrg.io. “Together, we are making AI more accessible and powerful so enterprises can generate better insights faster to achieve higher levels of efficiency and superior customer experiences. cnvrg.io Metacloud is the first AI platform in the industry to support CPU, GPUs and Habana Gaudis seamlessly in the same pipeline”
“Our combined solutions provide flexibility and high-performance processing, making AI more open, cost effective, and powerful”, said Eitan Media, Chief Business Officer at Habana Labs. “The ability to choose the Gaudi AI computational power—whether in on-premises infrastructure or in the recently launched Amazon EC2 DL1 instance– on the fly makes the advanced technology more available so enterprises can experience the benefits of AI/ML flexibly and with speed and ease.”
cnvrg.io Metacloud provides integrations with the Gaudi AI Accelerator, Intel CPUs, AWS EC2 instances, Azure, Google Cloud Platform, Dell, Red Hat, VMWare, Seagate and more. Exclusive early access to the cnvrg.io Metacloud is now available upon request on our website at https://metacloud.cloud.cnvrg.io/early-access.
cnvrg.io is an AI OS, transforming the way enterprises manage, scale and accelerate AI and data science development from research to production. The code-first platform is built by data scientists, for data scientists and offers unrivaled flexibility to run on-premise or cloud. From advanced MLOps to continual learning, cnvrg.io brings top of the line technology to data science teams so they can spend less time on DevOps and focus on the real magic – algorithms. Since using cnvrg.io, teams across industries have gotten more models to production resulting in increased business value. For more information, visit https://cnvrg.io/.