Tel Aviv — March 7, 2023 – ClearML, an open-source MLOps platform, today announced it has been certified to run NVIDIA AI Enterprise, an end-to-end platform for building accelerated production AI.
ClearML said the certification makes its MLOps platform more efficient across workflows, enabling optimization of NVIDIA GPUs. It also ensures that ClearML is compatible with and optimized for NVIDIA DGX systems and NVIDIA-Certified Systems from server manufacturers.
ClearML said its platform is designed to offer a little- to no-overhead solution for DevOps teams managing accelerated on- and off-prem computing infrastructures.
Installed within minutes, ClearML makes it easy to get up and running with AI. That ease of use is important given the growing importance of MLOps as more organizations recognize the need to automate and optimize their ML workflows to drive value and stay competitive. According to research conducted by ClearML in 2022, MLOps has achieved wide-scale adoption within companies and enterprises. The study found that 85 percent of respondents had a dedicated budget for MLOps in 2022, while an additional 14 percent of respondents saying they did not have budgets in place but expected to allocate funding for MLOps this year.
“We’re pleased to achieve this important certification from NVIDIA, as it will help our users and customers optimize their GPU loads so that their ML processes are even more efficient,” said Moses Guttmann, CEO and Co-Founder of ClearML. “This makes an organization’s accelerated computing power more accessible to their data science and AI teams, helping to maximize their hardware investment for machine learning.”
ClearML’s platform applies NVIDIA Multi-Instance GPU (MIG) technology, which enables teams to partition GPUs into as many as seven instances, which enables enterprises to match the correct computing power to each AI workload. This maximizes the efficiency of enterprise infrastructure and ensures that GPUs are maximizing power and avoiding idle units. Fractional and virtual GPUs created by ClearML can be accessed by containers, allowing different workflows to run in tandem on a single GPU.
As the operating system of the NVIDIA AI platform, NVIDIA AI Enterprise is an end-to-end, cloud-native suite of AI and data analytics software that’s optimized, certified, and supported by NVIDIA for production and support of applications built with the extensive NVIDIA library of frameworks and pretrained models. ClearML gives users and customers collaborative experiment management, powerful orchestration, easy-to-build data stores, and one-click model deployment. ClearML’s open source, frictionless, unified, end-to-end MLOps suite enables users and customers to focus on developing their ML code and automation, ensuring their work is reproducible and scalable.
“AI is enabling enterprises to capture new opportunities while reducing costs, and every organization is seeking tools that enable them to easily manage their AI production and deployment pipelines,” said Anne Hecht, Senior Director of AI software at NVIDIA. “With ClearML’s NVIDIA AI Enterprise certification, enterprises have an ideal solution for maximizing efficiency as they integrate AI into their operations.”
The new certification follows ClearML’s recent integration with software included in the NVIDIA AI Enterprise software suite. These include NVIDIA TAO Toolkit, a low-code AI model development for creating custom, production-ready AI models, which is helping customers better visualize results and intuitively compare experiments.
ClearML also works with NVIDIA Triton Inference Server software, and the company is working to integrate MONAI, a medical imaging AI framework started by NVIDIA and King’s College London. In early March, ClearML will be hosting a webinar with NVIDIA to discuss the collaboration and detail how NVIDIA AI Enterprise software, such as NVIDIA TAO Toolkit and NVIDIA Triton Inference Server, work together with ClearML’s MLOps platform.
